TABLE OF CONTENTS

COMMISSION STAFF WORKING DOCUMENT IMPACT ASSESSMENT......................... 6

1........... Introduction.................................................................................................................... 6

2........... Policy context................................................................................................................. 9

3........... Procedure and consultation of interested parties............................................................. 10

3.1........ Consultation of stakeholders and civil society................................................................. 10

3.2........ Studies and publications................................................................................................ 11

3.3........ Internal consultation, scrutiny of the impact assessment.................................................. 12

4........... Problem definition......................................................................................................... 13

4.1........ The general problems and their causes........................................................................... 13

4.1.1..... Insufficient protection of fundamental rights of suspected and accused persons............... 13

4.1.2..... Insufficient levels of mutual trust between Member States as a result of deficient protection of presumption of innocence..................................................................................................................... 16

4.2........ The specific problems................................................................................................... 19

4.2.1..... Non-respect of the right not to be presented as guilty by authorities before final conviction 19

4.2.2..... Non-respect of the principle that the burden of proof is on the prosecution and of the right of the accused to benefit from any doubt ("in dubio pro reo" principle)................................................................. 21

4.2.3..... Insufficient protection of the right not to incriminate oneself, the right not to co-operate and the right to remain silent.................................................................................................................................... 23

4.2.4..... Negative effects of decisions rendered in the absence of the person concerned at the trial ('in absentia')            25

4.3........ The scope of the problem............................................................................................. 27

4.4........ Baseline scenario: how would the problem evolve all things being equal?........................ 28

4.5........ Does the EU have power to act?................................................................................... 29

4.5.1..... Legal basis................................................................................................................... 29

4.5.2..... Subsidiarity: Why the EU is better placed to take action than Member States................. 29

5........... Objectives.................................................................................................................... 30

6........... Policy options............................................................................................................... 31

6.1........ Overview of policy options........................................................................................... 31

6.2........ Detailed description of the options................................................................................. 34

7........... Impact analysis of policy options................................................................................... 39

7.1........ Policy option 1 - Status quo.......................................................................................... 39

7.2........ Policy option 2 - Non-legislative action......................................................................... 40

7.3........ Policy option 3(a) - Legislative action............................................................................ 40

7.4........ Policy option 3(b) - Legislative option........................................................................... 43

8........... Comparative assessment............................................................................................... 45

9........... The preferred option..................................................................................................... 46

10......... Transposition, monitoring and evaluation........................................................................ 52

ANNEX I Summary of cited ECtHR case law............................................................................. 54

ANNEX II List of stakeholders consulted - on-line survey on presumption of innocence............... 62

Annex III Overview of the responses to the survey questionnaire.................................................. 64

ANNEX IV List of ECtHR cases in which EU Member States have been found in violation of presumption of innocence, between 1.1.2007 and 31.12.2012............................................................................................. 68

ANNEX V Legal situation in the Member States regarding presumption of innocence ('PoI')........ 69

ANNEX VII Examples of when lack of mutual trust can hinder judicial cooperation..................... 93

ANNEX VIII Glossary of main legal terms used.......................................................................... 96

Executive summary sheet

Impact assessment accompanying the Proposal for measures on the strengthening of the presumption of innocence in criminal proceedings

A. Need for action

Why? What is the problem being addressed?

There is insufficient protection of certain aspects of the principle of presumption of innocence of suspects and accused persons in the EU. This affects these persons' fundamental rights and their right to a fair trial, thus undermining mutual trust between judicial authorities which in turn hampers mutual recognition of judgements in criminal matters. The recognition and protection of presumption of innocence in the Charter of Fundamental Rights of the European Union ('the Charter') and in the European Convention for the Protection of Human Rights and Fundamental Freedoms ('the ECHR') have not prevented Member States from repeatedly violating this principle, in spite of the fact that level of safeguards in national legislation is, in general, acceptable. This allows the conclusion that the protection currently offered by the Charter and the ECHR is not enough to ensure that presumption of innocence is respected by EU Member States. Every suspect or accused person in criminal proceedings in any EU Member State may be affected.

What is this initiative expected to achieve?

The proposed measure aims ensuring that every suspect or accused person is presumed innocent in accordance with the Charter and the ECHR. This will lead to a higher mutual trust between the judicial authorities on which the European area of justice is built. It will set minimum standards of certain aspects of the right to be presumed innocent and furthermore, taking in due consideration the principles of subsidiarity and proportionality, appropriate remedies in case of breach.

What is the value added of action at the EU level?

The protection of the principle of presumption of innocence by the European Court of Human Rights ('the ECtHR') has not resulted in sufficient protection of suspects or accused persons in the EU. EU action would ensure the existence of common minimum standards and their effective protection in all EU Member States. In addition, the full panoply of redress mechanisms according to the Treaty (such as the duty to transpose directives; implementation monitoring by the Commission and the possibility of references for preliminary rulings) will be available to make sure that there was compliance with the right to be presumed innocent in criminal procedure contained in EU legislation.

B. Solutions

What legislative and non-legislative policy options have been considered? Is there a preferred choice or not? Why?

Status quo and non-legislative option have been considered but would not match the objectives. The latter encompasses awareness raising, drafting of guidelines, training, monitoring and evaluation systems and sharing of best practice (option 2). It could be implemented on its own or complement legislative options. Two legislative options were considered. Firstly, a Directive setting minimum standards in line with ECtHR jurisprudence completed by effective remedies in case of breach, for the following aspects of presumption of innocence: (1) right not to be pronounced guilty by the authorities before a final judgement; (2) burden of proof is on the prosecution (with certain limited exceptions) and any reasonable doubt benefits the accused; (3) right not to incriminate oneself, including the right not to cooperate and the right to remain silent (with certain well defined limited exceptions); (4) right to be present at one’s trial (with certain limited exceptions) (option 3(a)). Secondly, a Directive going beyond ECtHR standards of protection, with even more limited exceptions or no exceptions at all (option 3(b)).The preferred option is a combination of parts of options 2, 3(a) and 3(b). The preferred option fully respects the principles of subsidiarity and proportionality by proposing a differentiated level of EU intervention for each aspect of presumption of innocence, depending on several factors: (i) impact on the smooth functioning of mutual recognition instruments: particular attention should be given to those aspects which create concrete and tangible rights for the citizens – rather than general principles of procedural criminal law; (ii) stronger EU intervention is required for aspects which are not adequately protected by national laws and where problems go beyond the practical application of these laws, and (iii) stronger EU intervention is required for those aspects where ECtHR jurisprudence does not provide a standard which is sufficiently high in a common area of criminal justice (right to silence). The preferred option is as follows: for the first two aspects of presumption of innocence, it is part of option 3(a) (codification of ECHR standard, but no specific remedy in case of breach); the preferred option for the third aspect is a combination of options 3(a) and 3(b) since a specific remedy is justified (inadmissibility of evidence obtained in breach of this right) and furthermore no inferences from the exercise of these rights should be allowed; for the fourth aspect, option 3(a), including a specific remedy in case of breach, is justified. Implementation would be supported by horizontal measures on monitoring, evaluation and training (parts of option 2).

Who supports which option?

Some Member States (Ministries of Justice) are in favour of maintaining status quo (arguing that the ECtHR and its jurisprudence are sufficient and that their legislation already complies with the ECHR). Some others (DE, FR, IT, SI) are in favour of EU action if it proves necessary. Some stakeholders (defence lawyers, Bar Associations and academics) argue in favour of either non-legislative option or legislative option 3(a) or their combination, as they claim that the current situation is not satisfactory and since the strengthening of the principle is needed to promote mutual trust. Some individual stakeholders (defence lawyers, NGOs) support rather far-reaching option 3(b).

C. Impacts of the preferred option

What are the benefits of the preferred option (if any, otherwise main ones)?

There would be significant benefits by reducing the number of miscarriages of justice, thus promoting mutual trust of judicial authorities by increased clarification of Article 48 of the Charter. In the long term, the already limited financial impact estimated below should gradually tend to further reduce as the right to be presumed innocent should be more respected, and thus remedies for its breach would be less used. In addition, legislative options would bring overall reductions in current costs of ECtHR and domestic appeals, re-trials, financial compensation, aborted prosecutions due to breach of suspects’ fair trial rights resulting from an insufficient protection of the right to be presumed innocent.

What are the costs of the preferred option (if any, otherwise main ones)?

Given the lack of reliable data available figures are provided tentatively. In the most likely scenario, the costs of prohibiting inferences to be drawn from silence would be of 27 million euros per annum for 9 Member States altogether; for non-admissibility of use of evidence obtained in breach of the right not to cooperate, the estimated cost is between 7.500 and 75.000 euros per annum for 12 Member States altogether; for the right to be present at trial, the estimated costs are 523.00 euros per annum for 4 Member States altogether. An additional estimated amount of 1,3 million euros per annum is excepted, resulting from the monitoring system and reporting obligations to be fulfilled by Member States. No costs arise as regards the two first aspects of presumption of innocence.

How will businesses, SMEs and micro-enterprises be affected?

No direct or particular impact for businesses, SMEs and micro-enterprises is expected.

Will there be significant impacts on national budgets and administrations?

Non legislative action would lead to some financial costs that would be borne by national budgets of the Member States, except for a small part borne by the European Commission. In legislative options, the limited financial costs would be borne by budgets of the Member States.

Will there be other significant impacts?

For option 3(a) only minor legislative changes would be needed as some Member States would have to put in place more effective remedies in cases of breach of the rights already established in their legislation. Mainly the practical implementation and application of these rights will have to be ensured by the Member States. For option 3(b) more significant legislative changes would be needed.

D. Follow up

When will the policy be reviewed?

Specific empirical studies with an emphasis on data collection 3-5 years into the implementation of the initiative, to gain in-depth quantitative and qualitative insights into its effectiveness.

COMMISSION STAFF WORKING DOCUMENT

IMPACT ASSESSMENT

Accompanying the document

Proposal for measures

on the strengthening of certain aspects of the presumption of innocence and of the right to be present at trial in criminal proceedings

1.           Introduction

This impact assessment is for a measure on certain aspects of the right of suspects or accused persons[1] in criminal procedure to be presumed innocent until guilt has been legally established, in particular on the right not to incriminate oneself, including the right not to cooperate and the right to remain silent, and on the right to be present at trial[2]. The proposed measure aims at requiring the relevant authorities in the Member States to give the suspect or accused person enough procedural safeguards to exercise these rights effectively in accordance with the Charter of Fundamental Rights of the European Union and the European Convention for the Protection of Human Rights and Fundamental Freedoms. This will strengthen the right to a fair trial in the EU and ensure that the rights in other directives strengthening procedural rights will be effective. This will also, by benefiting the overall quality of justice in the EU, improve the mutual trust between Member States’ judicial authorities and thus facilitate a better functioning of existing EU legal instruments on judicial cooperation and on mutual recognition of judicial decisions in criminal matters between the Member States.

The presumption of innocence is a fundamental principle of human rights law broadly recognised, which lies at the heart of the notion of a fair procedure. It means that defendants are deemed innocent until proven guilty by court in a final judgment. Nevertheless, recent evidence[3] shows that this principle is not always respected in practice.

The principle of presumption of innocence is enshrined in all major international and regional instruments of human rights and fundamental freedoms: in Article 6(2) of the European Convention for the Protection of Human Rights and Fundamental Freedoms ('the ECHR')[4], in Article 48 of the Charter of Fundamental Rights of the European Union ('the Charter') as well as in Article 11(1) of the Universal Declaration of Human Rights[5]. It is also enshrined in the constitutions of the Member States, complemented by extensive jurisprudence, although in some Member States no detailed laws exist[6].

Presumption of innocence is one of the components of the right to a fair trial. It is a very broad principle and guidance on what precisely legally constitutes its content can be found in the case law of the European Court of Human Rights ('the ECtHR') which clearly set out 3 key requirements of this principle[7]:

– Public authorities including judicial authorities must not presume that the accused has committed the offence he is charged with;

– the burden of proof is on the prosecution and any doubt must benefit the accused (in dubio pro reo), and

– the prosecution must inform the suspect or accused of the case against him so that he may prepare and present his defence accordingly – this right has already been covered by Directive 2012/13/EU on the right to information in criminal proceedings and will therefore not be treated in this impact assessment.

The ECtHR has also expressly stated in its case law the existence of a clear link between presumption of innocence and other fair trial rights, in the sense that when such rights are breached presumption of innocence is inevitably also at stake:

– the right not to incriminate one-self, the right not to co-operate and the right to silence - the prosecution in a criminal case must seek to prove their case against the accused without resort to evidence obtained through methods of coercion or oppression in defiance of the will of the accused, and in this sense this right is linked to presumption of innocence[8];

– the right to be released pending trial[9] - Member States' rules on the right to be released pending trial / pre-trial detention do have an impact on the respect of presumption of innocence. However, pre-trial detention is already the subject of separate initiatives[10] and is therefore not covered by this impact assessment.

The right to be present at trial is also closely connected with presumption of innocence. As mentioned above, one of the key requirements of presumption of innocence is, as set out by the ECtHR, the right of the suspect or accused to be informed of the case against him; consequently, the right to be present at trial, or being able to waive such right after having been informed of it, is indispensable for the exercise of the right to be informed of the case[11]. If a suspect or accused is not given the opportunity to be present at trial because he was not informed of it, it is his right to be informed of the case that is also at stake and thus his presumption of innocence.

Some Member States argued during the experts meeting referred to in point 3.1 that 'in absentia' decisions are not linked to presumption of innocence. Such position is due to the fact that the ECtHR has not expressly treated 'in absentia' decisions as linked to presumption of innocence but rather to the general right to a fair trial of Article 6(1)[12]. However, the link with the right to be informed of the accusation and thus to presumption of innocence is undeniable.

In conclusion the following four aspects will therefore be treated in this impact assessment: · the right not to be presented as guilty by the authorities before final conviction; · the burden of proof is on the prosecution and the suspect or accused benefits from any doubt (in dubio pro reo); · the right not to incriminate oneself, the right not to co-operate and the right to remain silent; · the right to be present at one’s trial. The right to information, and the right to be released pending trial will not be treated, as initiatives to protect these rights have already been taken. Each of the four aspects that will be examined in this impact assessment  includes an important number of sub-issues (content of the right, exceptions, remedies) which do not all have the same importance in the context of the smooth functioning of EU mutual recognition instruments. This impact assessment will therefore, for reasons of proportionality and subsidiarity, concentrate on those sub-issues which are directly linked and are indispensable to the good functioning of these instruments. Moreover, for the same reasons, particular relevance will be given to those aspects which are not sufficiently covered by domestic law.

The existing principles established by case law of the ECtHR have proved not to be sufficient to achieve the necessary level of mutual trust between EU Member States required for the smooth functioning of the area of freedom, security and justice. In fact, in practice there still exist considerable shortcomings in the protection of this principle throughout the EU.

Presumption of innocence is overarching and complementary to other procedural rights. It is interconnected with and inseparable from other fair trial rights enshrined in Article 47 of the Charter and in Article 6(1) of the ECHR. Given the close link between those rights guaranteed by Article 6(1) of the ECHR and presumption of innocence, the ECtHR has even concluded that because there was a violation of Article 6(1), presumption of innocence was also breached[13].

The right to be presumed innocent might be interpreted in different ways with respect to natural and legal persons, in the light of the different needs and degrees of protection, as recognised in the case law of the Court of Justice[14]. As case law on the current applicable standards is more developed in relation to natural persons than in relation to legal persons, it is not possible, at this stage, to propose common minimum requirements on the right to be presumed innocent for legal persons. This impact assessment therefore only deals with presumption of innocence as regards natural persons.

This is in line with the "step-by-step" approach of EU intervention in the area of procedural rights in criminal procedure and the need for proportionate action. The need for future action in this field will be considered depending on the evolution of national legislation and of the case law.

2.           Policy context

The idea of justice is at the very heart of any democratic society. Approximation of fair trial standards across the EU will raise the awareness and confidence of citizens that their right to a fair trial is guaranteed throughout the EU - including when they exercise their right to free movement within a Union without internal borders - and by this it will address the issue of citizens' and national judicial authorities' lacking trust in the fair operation of another Member State's justice system.

Member States have agreed that mutual recognition should be the cornerstone of judicial cooperation, that is, that judicial decisions taken in one Member State should be considered as equivalent to each other wherever that decision is taken, and so enforceable anywhere in the EU. Judicial cooperation needs to be founded on mutual trust and confidence between the different judicial systems and the perception that the rights of suspects and accused persons are not respected in every instance has a disproportionately detrimental effect on mutual trust and, in turn, on judicial cooperation[15]. Thus, Article 82 of the Treaty on the Functioning of the European Union (TFEU) states that the principle of mutual recognition of judgements and judicial decisions should be facilitated by means of minimum rules on procedural rights.

In this context, the Stockholm Program[16] put a strong focus on the strengthening of the rights of individuals in criminal proceedings. In its point 2.4, the European Council invited the Commission to put forward six proposals contained in the Roadmap on Procedural Rights adopted under Swedish Presidency ("the Roadmap")[17], setting out a step by step approach to strengthening the rights of suspects and accused persons, the work on which is well advanced[18]. It was following the proposals of the Commission that a Directive on the right to interpretation and translation, a Directive on the right to information in criminal proceedings and a Directive on access to a lawyer in criminal proceedings have been adopted. On the side of strengthening victims' rights a Directive has also been adopted following a Commission proposal[19].

The purpose of the whole exercise of the Roadmap is to ensure the right to a fair trial. Presumption of innocence, together with its related rights, is part of that right. Various rights of suspects or accused persons in criminal proceedings established by the EU Directives over the past few years, such as right to interpretation and translation, right to information and right to access to a lawyer are not objectives themselves. They have a wider aim; they are rather tools to materialize the principle of the right to a fair trial. In case of persistent breach of the principle of presumption of innocence in the Member States, the overarching objectives of the measures already adopted under the Roadmap, including its key instrument which is the Directive on the right to access to a lawyer, would not be fully achieved.

It is for this reason that, in the Stockholm Program, the Council expressly invited the Commission to address the issue of presumption of innocence.

Once all envisaged initiatives on procedural rights in criminal proceedings are implemented, an environment of deeper mutual trust between judicial authorities will be in place.

3.           Procedure and consultation of interested parties

3.1         Consultation of stakeholders and civil society

Stakeholders were consulted on several occasions.

In 2006 the Commission published a Green Paper on the presumption of innocence, indicating what this principle covers, in line with the ECtHR case law[20]. At that time, 11 Member States replied to the consultation and, in general terms, were in favour of an EU initiative which takes into account what is really needed to reinforce the principle in practice. Independent experts and practitioners took the opportunity to point out an erosion of the principle of presumption of innocence and to underline, in particular as regards investigations against non-nationals or non-residents, that a principle of "presumption of guilt" seems to be more and more tolerated in national systems.

The Commission has also had regular contact with major stakeholders and has benefitted from consultations on the other initiatives attached to this package.

In the meeting of the Expert Group on EU Criminal Policy of 23 January 2013, the Commission had the opportunity to gather views from academics, practitioners, judges, defence lawyers and prosecutors.

Moreover, a meeting with Representatives of Ministries of Justice of Member States (including Croatia) specifically devoted to the presumption of innocence was held on 19 February 2013.

Member States were also directly consulted during the preparation of the external study referred to in point 3.2 below. However, it is not possible to say at this stage what the position of each individual Member States is. The result of such consultation and also of the Meeting on 19 February 2013 is that there is not an agreed official position yet, which depends on the concrete aspects of presumption of innocence covered by a possible proposal. Throughout this impact assessment the probable position of Member States has been inserted where possible.

An on-line survey was launched on 27 February 2013, published on the DG Justice website and included in the external study referred to in point 3.2 below. All major stakeholders were informed about this survey via e-mail[21]. The survey focused not only on the legal situation, but also on the functioning of the principle of presumption of innocence in practice. There was a total of 102 responses to the survey. It is not possible to know exactly from which 'category' of participant each response came from (lawyers, judges, prosecutors, NGOs, academics, etc.) because in some cases such information was not made available. However, the majority of responses came from lawyers or bar associations (around 70%). Three responses were also received from 'academics' and the other responses were from 'individuals'. An overview of the responses is available in Annex III.

Some Member States representatives and some other stakeholders (NGOs, lawyers, judges, prosecutors) have also taken part in workshops organized in five different Member States in the framework of the study referred to in point 3.2 below.

Given these different consultations, a formal open public consultation did not take place.

3.2         Studies and publications

This impact assessment relies on a number of studies and reports carried out from 2004 till the present date:

– An external "Study of financial and other impacts for an Impact Assessment of a Measure Covering the Right to be Presumed Innocent for Suspected or Accused Persons in Criminal Proceedings" carried out by the CSES ("Centre for Strategy and Evaluation Services") (hereafter the "CSES study");

– "Final Report – Towards a common evaluation framework to assess mutual trust in the field of EU judicial cooperation in criminal matters" – P. Albers, P. Beauvais, J.-F. Bohnert, M. Böse, P. Langbroek, A. Renier and T. Wahl, March 2013. This report presents a pilot project carried out in France, Germany and the Netherlands on the functioning of the European Arrest Warrant (hereinafter, 'EAW') system, including an analysis of the aspect of mutual trust between judicial authorities.

– A report presented in October 2012 by Fair Trials International on "Defence Rights in the EU"[22], based on an EU wide survey of lawyers and NGOs on the real barriers to a fair trial; it provides with a picture of the state of fair trial violations, including presumption of innocence.

– A report published by JUSTICE on ‘European Arrest Warrants – ensuring an effective defence’, 2012[23];

– "Effective Criminal Defence in Europe"[24], a comprehensive review of procedural rights in the EU, including the presumption of innocence, focusing on the situation in 8 Member States (and also in Turkey);

– "Effective Criminal Defence in Eastern Europe"[25], a study on procedural rights (including presumption of innocence) focusing on the situation in Eastern European countries, including Bulgaria and Lithuania;

– "Analysis of the future of mutual recognition in criminal matters in the European Union", an external study on Analysis of the future of mutual recognition in criminal matters in the EU finalised in 2008 conducted by the ULB (Université Libre de Bruxelles) and ECLAN (European Criminal Law Academic Network)[26];

– "Study of the laws of evidence in criminal proceedings throughout the European Union", The Law Society, 2004;

– On-going study "Inside Police Custody - an empirical Account of Suspects' Rights in Four Jurisdictions”, Maastricht University, University of the West of England, University of Warwick (co-financed by the Commission under the JPEN Program).

3.3         Internal consultation, scrutiny of the impact assessment

An Inter-service Impact Assessment Steering Group was created involving representatives from DGs COMP, EEAS, ENTR, ELARG, OLAF, HOME, CNECT, EMPL, REGIO, TRADE, HR / IDOC, the Legal Service and the Secretariat-General. The first meeting was held on 15 January 2013, the second meeting on 30 April 2013 and the final meeting was held on 30 May 2013. At those meetings and in subsequent communication with individual DGs, comprehensive feedback was received which has been taken into account throughout this report.

The European Commission's Impact Assessment Board (IAB) examined this report on 3 July 2013 and issued an opinion on 5 July 2013, in which it was requested to resubmit the report to the IAB, together with a number of suggested improvements to the same report.

A revised report, submitted to the IAB on 31 July 2013, took on board the recommendations of the IAB and introduced the following main modifications and clarifications:

-           Clarification of what the different aspects of presumption of innocence are and how they have been chosen to be included in this report;

-           Clarification of where the existing problems derive from;

-           Explanation on how the non-respect of presumption of innocence by Member States affects mutual trust between judicial authorities and thus the functioning of EU mutual recognition instruments;

-           Enhanced attention to subsidiarity and proportionality concerns, by better explaining the added value of the measure and by choosing a preferred option which foresees stronger EU intervention only as regards those procedural rights which: (i) are more related to the proper functioning of mutual recognition instruments; (ii) are not adequately protected at national level by reference to ECtHR case law; (iii) ECtHR case law does not provide for an adequate level of protection.

-           Better presentation of the costs, including a presentation of the costs of the preferred option for each Member State, and better explanation of stakeholders' views.

On 6 September 2013 the IAB issued a positive opinion on the revised report, together with some recommendations which are taken into account in the present final version of the report. The main modifications of this version, compared to the one on which the IAB issued its positive opinion, are the following:

-           Clarification of the need for action in relation to mutual trust, in particular by further explaining the baseline scenario and, more precisely, the extent to which other initiatives linked to this proposal are not sufficient to address the underlying problems;

-           Clarification of the rationale for action in relation to each aspect of presumption of innocence, and of a differentiated approach for action as regards each of those aspects;

-           Enhanced explanation on the assumptions underlying expected costs and benefits to national authorities as a result of the present proposal;

-           Better indication of the position of relevant stakeholders as regards the initiative.

4.           Problem definition

4.1         The general problems and their causes

1. Insufficient protection of fundamental rights of suspected and accused persons as a result of insufficient protection of the principle of presumption of innocence. 2. Insufficient levels of mutual trust between Member States as a result of insufficient protection of fundamental rights, which hampers the smooth functioning of mutual recognition of judgments and judicial decisions and judicial cooperation in criminal matters.

4.1.1      Insufficient protection of fundamental rights of suspected and accused persons

Description of the problem

There is currently no secondary EU law instrument that provides the right for suspected and accused persons to be presumed innocent throughout the entire criminal proceeding. There are, nonetheless, provisions on the presumption of innocence in Member States' constitutions and legal rules, in addition to extensive jurisprudence, which are on the whole more or less precise on the different aspects of the presumption of innocence.

An analysis of the legal situation in the Member States’ regarding protection of the presumption of innocence (Annex V, which contains detailed information on the system of protection of presumption of innocence in each Member State) shows that for some specific problems below, the legal regimes of the Member States appear to comply with the principles set out by the Charter and the ECHR, and it appears that it is the respect of the presumption of innocence in practice which is at stake. For other aspects, however, it is the legal regime itself which is insufficient. Indeed, although the general level of safeguards at national law level is in a general way acceptable, case law of the ECtHR shows that violations of presumption of innocence and its related fair trial rights have steadily taken place regardless of the fact whether these principles and rights are established by domestic law or not. According to the information available at the ECtHR database (Hudoc)[27], between January 2007 and December 2012, the ECtHR held for 26 times that there had been a violation of presumption of innocence under Article 6(2) by EU Member States[28].

It should be underlined that even if this figure already clearly demonstrates on its own the existence of a serious problem, we are strongly convinced that it is only the 'tip of the iceberg' and that violations of presumption of innocence by EU member States are significantly higher. Firstly, this figure only relates to violation of any of the three key requirements of presumption of innocence and does not include violations of other fair trial rights which are closely linked to it under Article 6(1) of the ECHR (such as the right not to incriminate oneself and the right to remain silent)[29]. Secondly, for various reasons (to be further developed hereafter, under the analysis of the causes of the problem) only a small percentage of those cases in which there would be grounds for an appeal to the ECtHR are indeed the subject of an appeal to the Strasbourg Court.

This allows the conclusion that the existence of certain principles established in the Charter and the ECHR and even in national constitutions and legislation, and further developed by the ECtHR and by national courts' case law, appears as insufficient to address, on their own, the shortcomings in the protection of the presumption of innocence. Even if the legal situation is prima facie satisfactory, there are still serious problems which are in general more linked with the operation of the presumption of innocence in practice. This was also confirmed by the field research carried out in the framework of the CSES study, which found that a number of stakeholders (NGOs, lawyers, judges) are aware of shortcomings in the practical observance of the presumption of innocence in other Member States[30].

Moreover, it should be borne in mind that the shortcomings of the protection of presumption of innocence also result in costs for EU Member States due to domestic and ECtHR appeals, re-trials, financial compensation and aborted prosecutions following the breach of the different aspects of this principle. By ensuring that fair trial rights are respected from the outset of proceedings, by enacting common minimum standards, one can avoid costs in the administrative and judicial system, costs which are usually not that visible[31]. By respecting fair trial rights and operating a system where there is trust in the respect of such rights, there are fewer appeals, fewer claims for retrial and one avoids appeals to and condemnations by the ECtHR. By having sufficient safeguards for fair trial rights, one also avoids challenges that obtained evidence is inadmissible. In mutual recognition proceedings, one avoids delays and costs arising therefrom, e.g. costs of providing pre-trial detention in EAW cases, or having more lengthy proceedings and more judicial and legal costs in case of non-consent in EAW cases.

Cause of the problem

This is partially due to the nature of the minimum rights set out in the Charter and the ECHR, which contain very general statements completed by the case law of the ECtHR. The case-law of the ECtHR is extremely linked to factual circumstances of each individual case and to the specificities of the legal system. For this reason, to rely on case law is problematic as it is by its nature piecemeal and reactive and results in very diverging interpretations by Member States. The ECtHR judgements only slowly build up a clear and consistent jurisprudence and may not even be followed by all national courts[32].

The European Criminal Bar Association, for example, has argued that it would be 'dangerous' to rely on ECtHR as the presumption of innocence standard because the perspective of the Court is ex post facto and it will take into account the entirety of the national trial proceedings[33]. There might be a non-violation case although there was a violation of the presumption of innocence in the first instance.

In addition the case law has so far not touched on the precise consequences of a violation of these rights, but has only decided case by case on establishing financial compensation in case of violation.

In several Member States only general and horizontal remedies are foreseen in legislation– in particular the right to appeal or the right to apply for civil law compensation in case of wrongful behaviour by the administration - but not specific remedies linked to the violation of the presumption of innocence and its related fair trial rights. Given the particularly heavy impact of imprisonment on the fundamental rights of the person concerned, this might not be satisfactory.

There are two main differences between a general right to appeal and a right to a retrial:

· as a general rule an appeal is limited to a re-examination of the case by a higher court in terms of questions of law and, contrary to a retrial, no new examination of evidence takes place. In contrast, a retrial means, as defined by the ECtHR, a procedure whereby a fresh determination of the merits of the charges is ensured[34], and this is not possible under the general rules of an appeal.

· an appeal intervenes only ex post, i.e., only after the first judgment has been delivered, which in relation to presumption of innocence means that the accused will only have the right to a fair trial in the higher court, and thus the right to have his case submitted to a double degree of jurisdiction – which is a fundamental general principle of law - does not exist in practice.

Relying on appeal as the only remedy to the breach of presumption of innocence and related fair trial rights affects citizens' fundamental rights, as there is an obvious difference between being in prison awaiting the result of the appeal proceedings and leading a normal life following an acquittal judgement delivered after the court has applied a specific remedy following the breach of presumption of innocence. When the right has been breached it is important to take any appropriate measure before the judgment has been delivered, and this is only possible with specific remedies in each case, such as to nullify the procedural steps taken following the breach of the right, declare evidence obtained in breach of presumption of innocence as non-admissible or, if these possibility are not allowed under national law, at least provide for the possibility of a retrial after the judgment has been passed.

Such shortcomings in the national enforcement systems cannot be compensated by the ECtHR enforcement mechanisms for three main reasons:

· The system of protection granted by the ECtHR is ex-post only. Ensuring justice in individual cases ex-post serves a different purpose from laying down generally applicable rules ex-ante and cannot be said to be equivalent.

· Moreover, the enforcement mechanisms of the ECtHR have not been sufficient to prevent EU Member States from too often violating the ECHR, in spite of the fact that they undertook to abide by the judgments of the ECtHR in any case to which they are parties (Article 46 §1 of the ECHR). Sixty per cent of the cases in which the ECtHR finds violations originate in failures to comply with the ECHR that have already been identified by the Court[35].

· The ECtHR's reluctance to lay down prescriptive requirements in these areas, which can be seen as a rationale for an EU measure. The approach of the ECtHR has not been especially activist in developing detailed and prescriptive rules in the area of Article 6(2) of the ECHR. It has left a margin of flexibility for presumption of innocence and related rights in light of the requirement to balance the fair trial rights of suspects or accused persons with the general public interest, as well as the diverse legal traditions of Member States. The court’s preferred approach is to set out generally expressed principles or minimum standards in its case law, to which contracting states are obliged to adhere pursuant to Article 53 ECHR.

Economic factors, such as the absence of legal aid, are also additional factors which can even more contribute to distort the proportionate relationship between the number of cases brought before the ECtHR and the actual number of cases where violations of presumption of innocence occur but are not brought to the ECtHR.

The insufficient protection of presumption of innocence across Member States is, in conclusion, detrimental to the protection of accused persons' and suspects' fundamental rights and to the general perception EU citizens have of the EU justice standard. A proper exercise of minimum rules at EU level on the various aspects of the presumption of innocence principle is essential in order to secure all other fair trial rights and to reduce the risks of a miscarriage of justice.

4.1.2      Insufficient levels of mutual trust between Member States as a result of deficient protection of presumption of innocence

The European area of justice in criminal matters has been built over the last 10 years on two types of instruments. First, the focus was on mutual recognition instruments, aiming at cross-border law enforcement. In a second stage the focus shifted towards instruments needed to balance the law enforcement aspects and designed to ensure that fair trial rights are preserved for individuals who are subject to intrusive procedures. Thus, from 2009, a number of procedural rights measures to safeguard the procedural rights of persons subject to cross-border investigation measures have been put in place.

It was after the 9/11 terrorist attacks in 2001, that EU action focused on facilitating law enforcement and the fight against crime. A series of instruments with the objective of prosecuting offenders were adopted. The most well-known instrument is the EAW which aims to expeditiously transfer suspects and accused persons between Member States, to ensure that the free movement of citizens across EU borders does not hamper effective cross-border law enforcement.  The instrument builds on the assumption that each Member State provides a system of justice which guarantees fair trial rights to a relatively similar degree.

To ensure fast-track and simple procedures for cross-border law enforcement and cooperation, the risk that the fair trial rights will not be respected in the ensuing main criminal proceedings once the suspect has been surrendered to the issuing Member State, does in principle not give the executing Member State a reason to refuse cooperation[36]. There is no express provision in the EAW that the executing Member State can refuse to cooperate because the legislation in the issuing  Member State does not respect fair trial rights, for example the right to remain silent, or the right not to incriminate oneself. It should however be noted that in recent years, a number of cases before the Court of Justice has touched upon the issue of whether there is a possibility to refuse to execute a mutual recognition request with reference to the risk of non-respect of fair trial rights in the ensuing criminal proceedings, something which demonstrates that there is an accruing need to foster mutual trust in the EU.

The Melloni case[37] shows that insufficient trust in the standards of protection of the presumption of innocence ('in absentia' judgements, in this case) may delay judicial cooperation (in this case the execution of a European Arrest Warrant). The national court refused the surrender of a person on the ground of different standards of protection in the requesting State. The case arrived at the European Court of Justice, which concluded that the difference in the standard of protection between the issuing and the executing Member State was no reason to refuse the surrender as long as certain minimums standards were respected. As a consequence, the person was surrendered, but only after a serious delay following several court proceedings. Even if in this case the EAW was ultimately executed, it serves to show how lack of mutual trust by Member Stes in one another’s standards can undermine the smooth working of judicial cooperation. The impacts of concerns in undermining mutual trust are illustrated by a recent English Appeal court case of Sofia City Court v Dimintrinka Atanasova-Kalaidzheiva[38]. UK court rejected twice an EAW against the same person on the grounds of abundance of evidence which casts doubts on the independence of the investigative and prosecuting process in Bulgaria. The appeal court doubted that a fair trial was possible in that particular case. If the right of a suspect to be presumed innocent only exists in theory, and is not effectively supported by all the different facets of criminal justice procedure, it will count for little in terms of engendering the necessary level of mutual confidence between Member States' judiciaries.

In addition to the EAW, a number of other measures have been adopted to facilitate cross-border law enforcement in the EU, for example the Convention on mutual assistance in criminal matters between the Member States of the European Union, the Framework Decision on the execution in the European Union of orders freezing property or evidence and the Framework Decision on the mutual recognition of confiscation orders in the European Union. These Decisions allow the authorities in one Member State (the issuing Member State) to ask the authorities in another Member State (the executing Member State) to take intrusive measures with regard to a suspected or accused person. Such measures can for example be collection of evidence e.g. by a house search, telephone tapping, by hearing a person, or freezing a person's asset in a bank account. They operate according to the same logic as the EAW, which means that the assumption is that the request from the executing Member State should be granted and mutually recognised. Hence, when State A (issuing Member State) asks State B (executing Member State) for an investigative measure affecting a person residing in that State for the purposes of conducting a criminal investigation that could lead to a criminal conviction, the State that shall execute the request has no guarantee or opportunity to verify that the underlying proceeding (in State A) respects the right to a fair trial, including the right of the presumption of innocence.[39] There are other EU mutual recognition instruments that are also affected by a lack of common minimum standards; EU instruments that aim to enforce the sentences and sanctions issued in other Member States.

To operate effectively, mutual recognition instruments must operate in a climate of mutual trust. When persons are subject to intrusive measures conducted in another Member State, it must be ensured that the integrity of the criminal procedure and the judicial authorities of that Member State fully respect the basic principles of the presumption of innocence and related rights of these persons, such as their right to remain silent and not to incriminate themselves. In cases of breach, there should be effective remedies to warrant that the position in the trial of the persons is not affected by violations of the principles and that they cannot be found guilty on the basis of evidence obtained on breach of these rights. The insufficient protection of these rights affects mutual trust negatively, something which in its turn undermines the confidence in cross-border instruments such as those referred to above. This is the logic underlying EU action with regard to procedural rights for suspects and accused persons, based on Article 82(2) of the TFEU. Fostering and reinforcing mutual trust by setting common minimum standards with respect to a set of procedural safeguards is indispensable to establish the climate of mutual trust which must underpin the proper working of the current mutual recognition instruments, as well as upcoming mutual recognition instruments such as the European Investigation Order.

Notwithstanding the influence of Article 6 of the ECHR in safeguarding and improving the right to be presumed innocent in Member States, it is clear that there are problems with the operation of the presumption of innocence and its related fair trial rights in practice in the different EU jurisdictions.

There is limited statistical quantifiable evidence on insufficient mutual trust between the Member States. Member States do not collect data on the number of judicial cooperation requests that are challenged or refused. Therefore, it is also difficult to quantify the problem.

Trust is based upon perceptions, and those problems described above consequently affect mutual trust and judicial cooperation. Clearly, any experience of lack of respect, or of poor respect, for human rights in the treatment of a citizen from one Member State in the criminal justice system of another Member State is potentially undermining. In this connection there are indications of judicial unease about divergent standards among Member States, as can be seen, for example, from the evidence of Lord Justice Thomas to the UK parliament’s Scott Baker inquiry concerning EAWs[40]. There are also indications of an absence of trust from other stakeholders, in particular defence lawyers, as demonstrated in the Final Report on a pilot project on the functioning of the EAW system (referred to in point 3.2 of this Impact Assessment), which concludes that 'About the aspects that are considered problematic or very problematic defence lawyers indicated in particular the quality of the judiciary/judges, the available capacity of the justice systems (judges, prosecutors), the right to a fair trial, the quality of legal representation and the conditions of detention'[41]. The right to a fair trial is therefore a concrete concern for stakeholders and plays a vital role in building mutual trust across the EU[42].

It can therefore be said that the lack of adequate protection of presumption of innocence and related fair trial rights also results today in insufficient trust between judicial authorities, which is detrimental to the mutual recognition of judicial decisions and other instances of judicial cooperation between Member States.

In practice, the system of mutual recognition often works sub-optimally as the swift operation is hampered by numerous challenges and appeals, resulting, as already pointed out in section 4.1.1, in additional costs and delays[43], partially due to long to complex and long drawn investigations into the systems of other Member States in such situations. Annex VII contains several examples of cases where insufficient trust in the respect of fair trial rights by another Member State caused such costs and delays. Ultimately this situation prejudices the resolution of a cross-border case for all parties involved, be it the suspected or accused, the victims or the general public.

4.2         The specific problems

As presumption of innocence is a broad principle, the general problem is the consequence of several specific problems. For the reason already set out in the introduction, this impact assessment concentrates on the four following specific aspects of the presumption of innocence and related fair trial rights, the protection of which is not sufficient within the EU.

4.2.1      Non-respect of the right not to be presented as guilty by authorities before final conviction

Public authorities, in particular law enforcement and judicial authorities, sometimes publicly refer to suspects or accused persons, in statements or in official decisions, as if they had been convicted of a crime before a court's final decision has been taken. Such behaviour harms the good reputation of the accused and can influence a jury or judge, who decides the case. Even if the person is acquitted afterwards by the court, he will be labelled as guilty by the general public and his life may be damaged forever, in particular if the case received broad media coverage.

In an important number of judgements, the ECtHR had the opportunity to set out a wide principle on the suspect's right to absence of public pronouncement of guilt before the final trial.

ECtHR principle: A court or public official may not publicly state that the accused is guilty of an offence if he has not been tried and convicted of it[44]. However, the authorities may inform the public of investigations and voice a suspicion of guilt[45], as long as the suspicion is not a declaration of the accused’s guilt[46] and they show discretion and circumspection. It should be noted that the scope of Article 82(2)(b) of the TFEU, referring to rights of individuals in criminal proceedings specifically, is not wide enough to allow an EU initiative to cover all public authorities under this obligation, but should be restricted to those who are directly involved in the proceedings, i.e. judicial authorities. This initiative will therefore only treat this aspect.

This principle derives from the general principle of presumption of innocence. It is protected by law at national level in all Member States, often by the constitution, even though not specifically laid down in the legislation of several of them (see table in Annex V, with an overview of the legal situation in the Member States regarding presumption of innocence). However, even in the latter cases it enjoys an implicit protection under the Member States’ case law or constitutional or criminal procedure provisions safeguarding the general principle of presumption of innocence.

As regards the remedies available, the ECtHR has not referred to any general principles or rules and has only decided case by case on financial compensation. Only 5 Member States have special rules for right of recourse (AT, FI, LT, PL, SE), whereas most Member States do not have specific remedies[47]. In the absence of such a specific remedy, some form of redress (such as appeal or financial compensation) is nevertheless available in all Member States. This is commonly because a public reference to guilt will constitute a violation of the suspect's right to be presumed innocent as set out in national law and Article 6(2) ECHR, or be a violation of the general procedural rights of the suspects or the general procedural duties of the court.

Despite established principles and general remedies throughout the EU, the right not to be referred to as guilty by judicial authorities is still breached within the EU. In spite of several benchmark cases in the 1990s' and early 2000, such as Allenet de Ribemont v. France, Daktaras v. Lithuania[48], Butkevicius v. Lithuania[49], several EU Member States have more recently been found in violation of this aspect of presumption of innocence (e.g. Pandy v. Belgium[50], Tendam v. Spain[51], Diacenco v. Romania[52], Poncelet v. Belgium[53], Lagardère v. France[54]).

Example:

In the ECtHR case Pandy v. Belgium[55], an investigating judge had referred to an accused of murder, in a public hearing, by comparing him to notorious serial killers. Such statements were reproduced in several articles in the press. The ECtHR found there had been a violation of presumption of innocence given that such statements involved a declaration of the accused's guilt which, firstly, encouraged the public to believe him guilty and, secondly, prejudged the assessment of the facts by the competent judicial authority.

In the ECtHR case Garlicki v. Poland[56] the facts were that a doctor specialising in cardiac surgery was arrested in the hospital where he was about to start a cardiac surgery by a dozen masked and armed officers of the Central Anti-Corruption Bureau (CAB). He was accused of medical negligence, harassment and receiving bribes from his patients. During a press conference, the head of the CAB referred to the defendant in the following terms: 'he is a ruthless and cynical bribe-taker. We have knowledge of several dozen bribes accepted by this doctor.' Furthermore, the Minister of justice made comments about the defendant that were deemed by the president of the constitutional court to have breached the constitution. The ECtHR recalled that this is a clear violation of presumption of innocence. However, in this case the court did not uphold the violation as all domestic routes for remedies were not exhausted.

Stakeholders have also confirmed the insufficient protection of this right in practice. Lawyers in Latvia mention that judges often openly express their attitude to the defendant in public, before the judgement[57]; lawyers in Spain refer that the main barriers to a fair trial as the absence of a real presumption of innocence and the breach of procedural safeguards[58]; discussions with stakeholders (lawyers and NGOs) in the frame of the CSES study for this impact assessment have highlighted that at least in the Netherlands, France and Poland, there are relatively regular breaches of the right not to be referred to as guilty by public officials (as opposed to judicial officials).

In conclusion, despite sufficient protection by means of adequate legal standards and general remedies in the EU Member States, a lack of respect of this aspect of presumption of innocence can still be observed in practice.

4.2.2      Non-respect of the principle that the burden of proof is on the prosecution and of the right of the accused to benefit from any doubt ("in dubio pro reo" principle)

Presumption of innocence presupposes that the burden of proof is on the prosecution, although it is admitted that in specific and limited cases it can be shifted to the defence. . Moreover, it also presupposes that any doubt on the guilt should benefit the suspect or accused person ('in dubio pro reo').

ECtHR principle: The members of a court should not start with the preconceived idea that the accused has committed the offence charged. The burden of proof is on the prosecution, and any doubt should benefit the accused (‘in dubio pro reo’)[59]. A court’s judgment must be based on evidence as put before it and not on mere allegations or assumptions[60].

The principle that the prosecution bears the burden of proof is however not absolute. In certain circumstances the suspect or accused may be required to bear a part of the burden of proof, e.g. to prove exculpatory circumstances in order to avoid being found guilty. In the case of Salabiaku v. France[61] , the Strasbourg court found that there was no Article 6(2) objection per se to the imposition of ‘strict liability’ in criminal proceedings for a customs offense (meaning that proof of certain objective facts alone is sufficient to prove guilt). However, it stressed that this should be applied “within reasonable limits which take into account the importance of what is at stake and maintain the rights of the defence”, in other words a presumption should be rebuttable (i.e., there should always exist a possibility to contradict a presumption of guilt) and a test of proportionality will apply in order to assess if such derogation to the general rule on the burden of proof is justified. One consideration would be the seriousness of the offence in question.

Furthermore, the principle that the burden of proof is on the prosecution is without prejudice to the power of initiative of the judicial authorities in the proceedings, for example as regards requesting new evidence to be produced.

Although theoretically one can put into question the fact that there are exceptions to the general rule, the standard set by the ECtHR is generally accepted as striking a correct balance between the public interest (the needs of prosecution) and the right of defence.

The ECHR principles seem to be rather well respected in the Member States' constitutions and legislations, in particular in the rather complex regulation of the admissibility of evidence in criminal proceedings and there is evidence that ECtHR case law is at the origin of national legislation or jurisprudence[62]. At least 13 Member States expressly admit the possibility of a shift of the burden of proof in some circumstances, but only in very limited cases can such shift of the burden of proof be questionable as regards compliance to the said ECtHR principles. In the Netherlands, e.g., as regards driving offences, an automatic presumption exists that the registered owner of the car has committed the offence, but this presumption is not rebuttable (contrary to the ECtHR case law) - the registered owner of a car is therefore not even allowed to produce evidence to prove his innocence and might consequently be convicted of a crime he has not committed.

As regards remedies, the situation also seems, as whole, satisfactory. In a minority of Member States the suspected or accused person whose right was breached can rely on a specific remedy leading to the nullity of the procedure, and in other Member States defendants can rely on the general remedies, such as to introduce an appeal (see table in Annex V, with an overview of the legal situation in the Member States regarding presumption of innocence).

There are however regularly cases of breach of this aspect of the presumption of innocence in the EU.

Example: A case widely commented on the press was the case of Mr Thomas Quick, in Sweden[63]. Mr Quick was the suspect in a case. He was accused of murdering 8 (or more) persons. He ‘confessed’ to his guilt and was duly convicted. Two years ago evidence came to light proving Mr Quick was in fact not guilty. The prosecutors, it transpired, had not made sufficient efforts to find all relevant evidence in the case. Instead, they only relied on the (false) guilt confession of Mr Quick, who was insane. This shows that the principle that the burden of proof lies on the prosecution and that reasonable doubts should benefit the accused was not respected.

Stakeholders have confirmed the insufficient protection of this right in practice. In Germany, NGOs criticise the fact that there is often pressure placed on suspects by the police to negotiate a plea bargain, which in practice undermines the principle that the burden of proof is on the prosecution[64]. In Hungary, practitioners indicate that court decisions ordering pre-trial detention often imply the Court's firm conviction about the suspect or accused; acquitting decisions seem to have to be substantiated in much more detail than convictions[65]; a significantly larger proportion of acquitting first instance decisions are quashed than convictions. In Bulgaria while the law provides clear and strong guarantees for the presumption of innocence, lawyers and some judges interviewed expressed concerns with an apparent accusatorial bias in the courts. They gave examples of individual cases where, in their view, a guilty verdict was delivered and eventually became effective even though insufficient evidence was presented to justify this[66].

In conclusion, despite sufficient protection by means of adequate legal standards and general remedies in the EU Member States, which seem to be in conformity with the ECtHR principles, a breach of this aspect of presumption of innocence still occurs too often in practice.

4.2.3      Insufficient protection of the right not to incriminate oneself, the right not to co-operate and the right to remain silent

Although not explicitly included in the ECHR, the ECtHR has recognised that the right not to incriminate oneself, the right not to co-operate and the right to remain silent are "generally recognised international standards which lie at the heart of the notion of a fair procedure"[67]. Without these immunities, the person could be improperly forced to produce evidence, and hence the principle that the burden of proof is on the prosecution would not be respected. By forcing someone to confess a crime, a suspect or accused might be found guilty of a crime he has simply not committed, or by forcing someone to confess a certain version of the facts under investigation, exculpatory circumstances might not be taken into account.

These principles have been clearly set out in the case law of the ECtHR.

ECtHR principle: The presumption of innocence includes the right not to be compelled to testify against oneself and not to confess guilt, the right not to cooperate and the right to remain silent. The rationale of these rights lies, inter alia, in the protection of the accused against improper compulsion by the authorities, thereby contributing to the avoidance of miscarriages of justice and to the fulfilment of the aims of Article 6 of the ECHR[68].

The ECtHR accepts, however, that in certain limited circumstances neither drawing adverse inferences from a person’s silence nor compelling the defendant to give evidence existing independently of the will of the accused necessarily infringes the presumption of innocence. In both situations, factors to which the Strasbourg court will have regard in determining whether there has been a violation include the nature and degree of compulsion, the weight of the public interest in the investigation and punishment of the offence at issue, the existence of any relevant safeguards in contracting states’ law and the use of the material so obtained in subsequent proceedings[69]. As a consequence, the privilege does not extend to material which may be obtained from the accused through the use of compulsory powers but which exists independent of the will of the suspect such as, inter alia, documents acquired pursuant to a warrant, breath, blood and urine samples and bodily tissue for the purpose of DNA testing[70].

The exceptions to these principles as established by the ECtHR are often put into question by stakeholders, in particular defence lawyers and NGOs as regards inferences being drawn from the exercise of these rights. They opine[71] that the suspect and accused's right to silence should be absolute and unqualified. They claim that, derived from the right to human dignity, this right would be illusory if the accused had to fear that his silence will be used against him later in the criminal proceedings.

On the other hand, it is generally accepted that the use of compulsory powers in obtaining evidence which exists independently of the will of the accused does not infringe the presumption of innocence and is justified by the public interests of prosecuting crime, as long as it does not violate other rights such as the prohibition of torture of Article 3 of the ECHR.

The ECHR principles are generally enshrined in the Member States' constitutions and legislation. There are, however, variations among EU jurisdictions as regards the nature and the scope of the circumstances in which the law permits inferences to be drawn, from relatively wide to more limited[72]. In some Member States, refusal to co-operate with the prosecution can lead to adverse inferences and / or be taken as incriminatory evidence (in Belgium, Cyprus, Finland France the UK, Ireland, Latvia, the Netherlands or in Sweden) - see table in Annex V, with an overview of the legal situation in the Member States regarding presumption of innocence.

As results from the same table in Annex V, there is also a lack of effective and specific remedies for breaches of the right to remain silent and right not to cooperate in some Member States (BE, BG, CY, EE, ES, HR, IE, LT, LV, NL, PL, UK, SE). For the right to silence, remedies are available during the trial proceedings in some EU jurisdictions (e.g., evidence can be declared inadmissible or the case can be dismissed), which is the most efficient way to recover the breach; whereas in a majority of Member States the remedy, often in form of a general right of appeal, is only available after the judgement of the first instance trial. An appropriate specific remedy could indeed be, e.g., a provision by which an unfairly obtained statement should be excluded from the evidence to be assessed by the court[73]. The principle of free evaluation of evidence by the Court, which is a principle generally recognized in all EU Member States, should not mean an absolute principle of free admissibility of all available evidence and should nevertheless allow excluding from the case evidence obtained in violation of fundamental rights, which seems to be in terms of legal certainty the correct means to ensure that the judge is not influenced by such evidence when taking the final decision.

Furthermore, there is clear evidence of breaches in practice of the right not to incriminate oneself, the right not to cooperate and the right to remain silent in Member States, as several times pointed out by the ECtHR.

Example: When Mr Jalloh[74] was going to be arrested in Germany on the street under suspicion of being a drug dealer, he swallowed a little bag believed to contain drugs. The public prosecutor ordered that emetics be administered to Mr Jalloh by a doctor in order to provoke the regurgitation of the bag. He was taken to a hospital and, given that he refused to take the medication necessary to provoke vomiting, he was held down and immobilised by four police officers. The doctor then forcibly administered to him a salt solution and an emetic through a tube introduced into his stomach through the nose. In addition, the doctor injected him with apomorphine, another emetic that is a derivative of morphine. As a result, the applicant regurgitated one little bag containing 0.2182 grams of cocaine. Besides concluding that Mr Jalloh had been subjected to inhuman and degrading treatment contrary to Article 3 of the ECHR (on prohibition of torture), the ECtHR also held that allowing the use at Mr Jalloh's trial of evidence obtained by the forcible administration of emetics infringed his right not to incriminate himself and therefore rendered his trial as a whole unfair.

Stakeholders have also confirmed the insufficient protection of the right to silence in practice, e.g. where police authorities suggest to suspects or accused that if they exercise their right to silence the courts will draw adverse inferences from this and/or they are more likely to be detained before trial proceedings (lawyers from EE and IT). Improper pressure is often used in order to convince the suspect or accused to cooperate. E.g. practitioners in Austria have mentioned cases where a person is informed about the right to remain silent, but at the same time is advised not to use it as it could be seen as an aggravating circumstance in the criminal proceedings. In the Netherlands police often seek to undermine a decision of the suspect to use his right to remain silent[75].

To summarise the above: First, the protection of these rights in domestic law of Member States is not sufficient given that inferences are drawn from the silence, i.e., exercising the right to silence can in certain circumstances be used as incriminatory evidence. Second, in some EU Member States there are no specific remedies available for breaches of these rights and the only available remedy is the right to appeal, which in this case is not satisfactory, as a general right to appeal does not exclude from the file such illegally obtained evidence, which does not prevent the Court to be influenced by it. Third, breaches of the right can often be seen in practice.

4.2.4      Negative effects of decisions rendered in the absence of the person concerned at the trial ('in absentia')

If a person is not present during the hearing, it is his right of defence that is at stake. The defendant will in that case neither be able to give his version of the facts to the Court, nor will be able to present evidence accordingly, and might therefore be found guilty without enough grounds for such a conviction.

ECtHR principle: The right to be present at the trial is linked to the right to be informed of the accusation, so that the accused may prepare and present his/her defence accordingly.

A trial 'in absentia' is compatible with the ECHR as long as the accused, if he has not waived his right to present, may subsequently obtain, from a court which has heard him, a fresh determination of the merits of the charge where it has not been established that he has waived his right to appear and to defend himself[76]. If the suspect has received a summons and deliberately does not attend the trial, a retrial may be refused. A lawyer who attends a trial for the apparent purpose of defending the accused in his absence must be given the opportunity to do so. In appeal or cassation proceedings the suspect or accused’s right to be present can be restricted if the proceedings are limited to questions of law and do not review the facts. The requirement to hold an appeal hearing in public and in the presence of the accused depends, in summary, on the nature of the appeal system, the scope of the court of appeal’s powers and the manner in which the applicant’s interests are presented and protected.

The respect of the right to be present at trial therefore still very much depends on national law, which present an important degree of variety. While some Member States allow accused persons to waive their right to be present at their trial and this is considered sufficient by the ECtHR principles, in other Member States the presence of the defendant appears to be mandatory in practice for more serious offences (Ireland, Cyprus and, to a lesser extent, Germany).

The situation becomes more serious when certain specific situations clearly justify intervention, in view of a clarification of what should be the minimum standard for a case to be judged 'in absentia'. E.g. in Finland, 'in absentia' decisions are possible if the defendant has been informed about the possibility that the case may be judged without his/her presence (without having to give consent to the court to be judged 'in absentia'), and if it is deemed that his/her presence is not necessary. The defendant may be sentenced to a fine or to imprisonment for a maximum of three months (see table in Annex V, with an overview of the legal situation in the Member States regarding presumption of innocence).

It is striking that national laws on remedies are not currently compliant with the ECtHR findings in the ECtHR case Colozza v. Italy, that trials 'in absentia' must provide for the accused subsequently to obtain, from a court which has heard him, a fresh determination of the merits of the charge where it has not been established that he has waived his right to appear and to defend himself[77]. Only if the suspect has received a summons and deliberately does not attend the trial, a retrial may be refused. It appears that in at least 4 Member States (BE, BG, HU and LV) the opportunity for a retrial is not guaranteed in these circumstances, i.e. in cases where the accused has not waived his right to be present – the person either is entitled only to appeal the decision, or he has no right beyond those he would have ordinarily (including, potentially, appeal) have had if he had attended the trial in person[78].

Example: In 2005, Mr Chen was arrested in the UK on the request of Romania where he had been tried in his absence in 1995 and given a 20 year prison sentence. Mr Chen insists that he knows nothing about the alleged offence and was in fact in Hungary on the date in question. Mr Chen had no knowledge that his trial was taking place and was unable to present any evidence, but in 2006 the UK courts ordered his extradition to serve the 20 year sentence. They did this based on the assumption that Romania would in practice allow him a retrial. Following his extradition, Mr Chen applied for a retrial but no retrial was ordered. Mr Chen’s Romanian lawyer has recently filed an application to the ECtHR.

The enormous importance of 'in-absentia' judgements on mutual trust is confirmed by the existence of a dedicated EU instrument. Council Framework Decision 2009/299/JHA (in absentia)[79], which sets out common minimum standards to avoid that a Member State refuses to recognise and to execute judicial decisions issued by another Member State for the mere reason that the suspected or accused was not present at his or her trial. Mutual recognition can only be refused when the standards in the issuing Member State do not comply with those set out in the Framework Decision.

By its very nature, however, this 'third pillar' instrument is not sufficient to ensure the respect of the ECHR principle throughout the EU. Strictly speaking, the framework decision does not oblige Member States to respect the common minimum standards in all national proceedings, but it merely defines a possible ground of refusal for cases of judicial cooperation if these common minimum standards are not respected.

To summarise, it is clear from the above that not only there are important variations in the level of protection among Member States, but that several Member States do still not comply, in their legislation, with the ECHR principles, be it for the conditions of 'in absentia' trials or for the remedies.

4.3         The scope of the problem

In 2010, there were almost 10 million criminal proceedings (only serious offences and misdemeanour) in all EU Member States[80]. Potentially all suspects and accused persons in criminal proceedings in the EU are affected by these problems.

Member States currently do not collect data on the number of proceedings in which insufficient protection of presumption of innocence is complained about or has led to judicial decisions being appealed and upheld or reversed by a higher court[81]. Nevertheless, recent cases such as Garlicki v. Poland (described in section 4.2.1) or case of Mr Chen (described in section 4.2.4) may help illustrate the potential scale and impact of these types of problem.

The number of cases in which the ECtHR found a Member State of the EU in breach of rights covered by Art 6 § 2 ECHR has been constant over the last five years (26 cases from 2007-2012[82]). At least 10 Member States, some of them repeatedly, were found by the ECtHR in violation of the right to be presumed innocent over the past five years. Particularly striking is the fact that this fundamental right, which forms an essential part of the right to a fair trial, would be expected to be observed by all EU Member States in the 21st century.

The number of cases reaching the ECtHR are, as demonstrated in the section 'cause of the problem', under point 4.1.1, contingent on a variety of factors which do not all relate in proportion to the actual number of cases where violations of presumption of innocence may have occurred. The number of cases brought before the ECtHR and the actual number of cases where violations of presumption of innocence occur but are not brought to the ECtHR can be distorted by a number of factors (see point 4.1.1). In any event, the problem is not essentially one of numbers. A single well-publicised case is enough to impact judicial trust and cooperation. There is no evidence that such cases are becoming rarer.

4.4         Baseline scenario: how would the problem evolve all things being equal?

Whilst certain European principles on presumption of innocence have been established in the past years, namely by the ECtHR, this has, however, not resulted in sufficient protection of suspected and accused persons everywhere in the EU according to legislation and practice of Member States.

Further changes on the basis of the existing legal framework are unlikely. Contacts with Member States during the preparation of this Impact Assessment have not shown any legislation in preparation. On the contrary, fast track procedures and summary procedures have increased in recent years as a result of cost-saving efforts[83], and there is a serious risk that they will develop in such a way as to further undermine the respect of this principle in the future if no action is taken.

Also the Charter itself does not provide a satisfying solution. While it binds the EU institutions and bodies in all instances, Member States are the addressees of the Charter only when implementing Union law. This entails that, in the absence of EU law, Member States are not bound by Articles 47 and 48 of the Charter in the conduct of criminal proceedings. In such cases individuals may not invoke these provisions, either directly or indirectly, in order to challenge the infringement of their rights by domestic institutions.

Through the Roadmap, the EU has already taken action to improve fair trial rights which will have to be implemented by the Member States in the coming years. Although this will have some impact on the protection of the presumption of innocence, this is not sufficient: for example, in the Directive 2012/13/EU on the right to information in criminal proceedings, the right to be informed of the right to remain silent is guaranteed in Article 3(1)(e); but the said Directive does not itself provide suspect or accused persons with the right to remain silent, but merely with the right to be informed about such right, such as regulated in national law. Directive 2013/48/EU on access to a lawyer also brings a positive effect as the presence of a lawyer is per se an important safeguard of all procedural rights, including presumption of innocence.

Furthermore, the forthcoming initiatives on legal aid (presented as a package with this proposal) will ensure that the right to access to a lawyer becomes effective in practice also for persons who don't have sufficient means to afford a lawyer.

However, the first three aspects of the principle of presumption of innocence which are covered by the present initiative have not been dealt with at all in the measures adopted or proposed. As regards the right to be present at one’s trial, the Framework Decision on 'in absentia' proceedings safeguards the right of a person to be present at a trial and sets out conditions for exception from this right. However, as explained under 4.2.4, it establishes this right only indirectly by formulating possible ground for refusal in judicial cooperation. This Framework Decision is only applicable in criminal proceedings subject to the mutual recognition principle, such as in EAW proceedings. It is desirable that the provisions of the Framework Decision 'in absentia' are extended to all criminal proceedings without difference, as it is not always clear at the beginning of the proceedings whether any cross-border element will be involved at a later stage.

4.5         Does the EU have power to act?

4.5.1      Legal basis

The EU's legislative competence for a Directive laying down minimum rights in criminal procedure is set out in Art 82(2)(b) TFEU[84]. Pursuant to this provision, minimum rules concerning the rights of individuals in criminal procedure may be adopted by means of directives, to the extent necessary to facilitate mutual recognition of judgments and judicial decisions and police and judicial cooperation in criminal matters having a cross-border dimension. The necessity for action at EU level is demonstrated by the demonstrated lack of trust among Member States’ authorities which is detrimental to the smooth functioning of mutual recogintion and judicial cooperation.

Art 82(2)(b) TFEU, provides the legal basis for legislation applicable not only to cross-border criminal proceedings (i.e. proceedings with a link to another MS or a third country) but also to domestic cases as a precise, ex ante categorisation of criminal proceedings as cross-border or domestic is impossible in relation to a significant number of cases. All previous Commission Proposals for Directives on procedural rights of suspect and accused persons followed this logic as they set up minimum standards for certain rights in all criminal proceedings, not only in cross-border criminal proceedings.

4.5.2      Subsidiarity: Why the EU is better placed to take action than Member States

It is considered that there is a need for EU action based on the following factors:

(a) Enhancing mutual trust between judicial authorities (see section 4.1.2): The EU is establishing its own, unique system of judicial cooperation in criminal matters having a cross-border dimension: such a novel system is based on the principle of mutual recognition throughout the EU and calls for a guarantee of uniform minimum standards of fundamental procedural rights protection in the EU which contribute to mutual trust between judicial authorities that such rights are effectively respected. The problem has a cross-border dimension because if certain Member States do not respect the rights, this will have an impact in other Member States and ultimately in the smooth functioning of mtual recognition legal instruments. EU action will therefore be focused in those aspects which are directly linked to the functioning of mutual recognition instruments and to police and judicial cooperation having a cross-border dimension.

(b) Movement of persons: Persons can be involved in criminal proceedings outside their own EU Member State and the needs of those suspected and accused persons need to be tackled at EU level. In the European Union people are constantly travelling and moving across borders. Around 13.6 million Europeans reside permanently outside their home country, 10% of Europeans have lived and worked abroad during a period of their lives and 13% have gone abroad for education or training[85].

These figures show the importance of ensuring proper, effective action on the rights of those who get involved in criminal proceedings, in their own country or while travelling or living abroad. The EU must ensure that suspects and accused persons benefit from a level playing field. They may not be fully aware of various specific aspects of the procedure but they should anyway be confident of getting a fair trial anywhere in the EU, including the protection of the right to be presumed innocent. The Charter contains such right; however, it can only be invoked in an individual case if the matter is related to the application of EU law by the Member State. In the absence of EU law, Member States are not bound by Articles 47 and 48 of the Charter in the conduct of criminal proceedings and individuals may not invoke these provisions, either directly or indirectly, in order to challenge the infringement of their rights by domestic institutions.

(c) Limits of the ECtHR enforcement mechanisms: As demonstrated before (see section 4.1.1 supra), the ECtHR enforcements mechanisms are not sufficient to ensure that the ECHR standards are applied in practice throughout the EU, as they have not prevented EU Member States from repeatedly violating Article 6(2) of the ECHR, in spite of the fact that they undertook to abide by the judgments of the ECtHR in any case to which they are parties (Article 46 §1 of the ECHR). To ensure an effective compliance of presumption of innocence by the Member States, EU redress mechanisms should therefore be available. Once and if the EU takes legislative action, the full panoply of redress mechanisms according to the Treaty (such as the duty to transpose a directive into legislation in the Member State; implementation monitoring by the Commission and the possibility of references for preliminary rulings) will be available to make sure that there was compliance with the right to be presumed innocent in criminal procedure contained in EU legislation. In addition, an EU directive would be applicable (even despite the absence of timely transposition, under the doctrine of direct effect) before domestic courts and would take precedence, under the principle of primacy of EU law, over conflicting domestic provisions. Risks of violation of EU standards by national authorities would be diminished by the mechanism of reference for a preliminary ruling, which allows the ECJ to provide the domestic court with the correct interpretation of EU provisions, in the course of (and not after) national proceedings.

Bringing the principles of the ECtHR case law into EU law will therefore also in itself improve legal certainty for citizens, who will be able to rely on the provisions of an EU instrument establishing minimum rules applicable in all Member States, rather than in such general principles which derive from concrete cases and which are not necessarily followed and enforceable in all EU countries.

5.           Objectives

Objectives:

General: || · To guarantee for EU citizens an effective high-level standard of protection of fundamental procedural rights in criminal procedure. · To enhance mutual trust thus facilitating mutual recognition of judgments and judicial decisions in the EU and improving judicial cooperation in the EU.

Specific: || · To ensure that all suspects or accused persons are presumed innocent during the entirety of criminal procedure until proved guilty according to law, and treated as such by Member States' judicial authorities. · To ensure that authorities dealing with judicial cooperation and involved in the execution of a criminal sanction, of an investigation  measure or of a European Arrest Warrant issued in another Member State are confident that the underlying decision was taken in full respect of the principle of presumption of innocence.

Operational: || · To ensure that no suspect or accused person in the EU is presented by any public authority as guilty before a final judgement. · To ensure that the burden of proof of the culpability of any person suspected or accused in the EU is on the prosecution and that any doubt shall benefit that person. · To ensure that the right to remain silent, the right not to cooperate and  privilege against self-incrimination of suspected and accused persons confronted to criminal justice systems in the EU are duly protected at any stage of the procedure. · To ensure by that the judgment is taken in the presence of the accused person, except in specific cases (“in absentia decisions”).

The present initiative forms part of a package of measures for improving mutual trust. Only once all the measures envisaged in the Roadmap and the Stockholm Programm are in place will it be possible to achieve the general objective. The following options are assessed against the specific and operational objectives above.

6.           Policy options

The options for addressing the problems as defined in part 4 of this impact assessment, in line with the objectives as established in part 5, are set out below.

In accordance with Communication from the Commission on the Strategy for the effective implementation of the Charter of Fundamental Rights by the European Union[86], this impact assessment examines the impact on the Fundamental Rights of the options proposed, in particular in the light of the 'fundamental rights check list' presented in the Communication.

For all the policy options, reference is made to suspects or accused persons as the category of persons who would be affected by these options. The phrase "suspects or accused persons" – consistently used in EU policy documents and previous legislation in this area - encompasses all people who are involved in criminal procedure, against whom a suspicion that they have committed a criminal offence exists, irrespective of the terms used in domestic law[87]. The breadth of the phrase is such that it does not require a definition, which would be very complex and difficult to square with national definitions. In accordance with the principle of proportionality of EU action, no rules should be made when no need for them can be shown.

6.1         Overview of policy options

We have considered four options: retention of the status quo (option 1), a soft law option (option 2) and two legislative policy options (options 3(a) and 3(b)). The retention of the status quo would involve taking no action at EU level, while the other three alternative policy options will improve, to a different extent, the protection of the right to be presumed innocent for suspects and accused persons across the EU.

Option 1 Status quo || Retention of the status quo. This option would involve taking no action at EU level.

Option 2 Soft law measure obligation || Non-legislative action (soft law). This option would include guidelines and training on good practice on suspects' or accused persons' right to be presumed innocent. Option 2 could be implemented on its own or as a first step or supporting action for implementing EU legislation.

Option 3 Legislative measure || a) Directive setting minimum rules on the principle of presumption of innocence applying the ECtHR acquis and establishing appropriate specific remedies in case of breach

b) Directive as under option 3(a), but setting common minimum rules which provide for a higher level of protection than the ECtHR acquis as regards each of the specific problems identified above (except for the absence of public references to guilt before conviction, where it is not possible to go beyond the ECtHR principle). This would be done by further limiting or even excluding the possibility of having exceptions to the general principle.

Relation between problems, objectives and policy options

Problem definition || Causes of the problem || Specific problems || General objectives || Specific objectives || Operational objectives || Policy options

|| General nature of the minimum rights set out in the Charter and in the ECHR || Non-respect of absence of public references to guilt || || || Ensure that no one is presented as guilty by judicial authorities before a final judgment || Status quo

Insufficient protection of fundamental rights as a result of insufficient protection of PoI || ECtHR case law is linked to individual circumstances of the case and can lead to divergent interpretations || Burden of proof on prosecution and any doubt should benefit the accused || Protection of fundamental procedural rights in criminal law || Ensure that suspect or accused are presumed innocent and treated as such during the procedure || Ensure that the burden of proof is on the prosecution and any doubt benefits the accused Ensure the protection of the right not to incriminate one-self, the right not to cooperate and the right to remain silent || Non legislative option

|| ECtHR has not touched on the consequences of the breach of PoI || Insufficient protection of the right not to incriminate oneself, the right not to cooperate and the right to silence || || Ensure that authorities involved in judicial cooperation are confident that the underlying decision respected PoI || Ensure that, apart from specific cases, the judgment is taken in the presence of the person || Legislative option (ECtHR principles + remedies)

Insufficient levels of mutual trust

enforcement mechanisms of the ECtHR are not enough

Lack of adequate protection of PoI

Right to be present at one’s trial

Enhance mutual trust

Legislative option (above ECtHR principles + remedies)

6.2         Detailed description of the options

Policy option 1, the status quo, has been presented in the baseline scenario (see Section 3.5). Some Member States were in favour of maintaining status quo. However, only a minority of other stakeholders was in favour of the status quo option as they feel that action needs to be taken in respect of presumption of innocence.

The other policy options are described in the table below and meet to varying degrees the objectives outlined in Section 3 above.

Policy option 2: Non-legislative measures are common to all four aspects of presumption of innocence.

These would include:

· Drawing up non-statutory guidance for staff in competent authorities on the respect of presumption of innocence and on the way in which the right to be presumed innocent can be safeguarded.  These would be drawn at experts meetings summoned by the European Commission and would include representatives of all Member States, as well as experts in the field of procedural criminal justice;

· Exchange of best practice which would take the form of workshops organised in groups of Member States with similar judicial traditions. These workshops would allow practitioners to share their experience and best practices. Sharing information on Member States’ legal systems could also be done through exchange and training schemes where judges spend some time in different EU jurisdictions to learn and understand the way in which the principle is approached in different Member States.

· Training – even in the absence of a legislative option, training of magistrates and prosecutors could take place and play a role in enhancing the right to be presumed innocent.  As discussed in the subsequent sub-sections, in some cases, the breach of the presumption of innocence resulted from carelessness on the part of the authorities that could have been prevented if more robust training had been provided;

· Improved monitoring - monitoring of the system safeguarding the presumption of innocence, and some uniform means of collecting data on cases where a breach of the presumption of innocence has been experienced.  Monitoring exercises could also help to collect better information on the scope of the problem and inform any future policy decision on the need for intervention, beyond that identified in this impact assessment.

The guidelines would be drawn up during a series of meetings convened by the European Commission and involving national experts from all the Member States. We have assumed that three day-long meetings would be necessary to agree on the guidelines. The guidelines  would be then disseminated amongst stakeholders. In addition, there would be training programmes to disseminate the best practices identified.

We then propose to group Member States according to their legal cultures, traditions and systems to organise workshops where the issues will be discussed. In those meetings, where different representatives of the legal profession would be invited (judges, prosecutors, lawyers, law enforcement authorities etc), discussions would include current practices in each of the represented systems and the identification of best practice. The output of those meeting would be a list of best practices, ranked by the ease in which they could be put in place.

For most of the issues, identified, a system of training of different actors of the criminal procedure (police, lawyers, prosecutors, judges etc) would be beneficial.  For example as regards the absence of public references to guilt training at national level could include sessions to help legal officials develop linguistic formulae for discussing the suspect or accused and the relevant criminal proceedings without the court being seen to imply (or expressly state) the suspect’s guilt before any final verdict. As regards the burden of proof, training would include sessions to ensure legal officials are aware of the types of cases where the burden of proof can be reversed and how a breach of this right should be remedied during the procedure. 

Finally, encouraging Member States to set up a monitoring system would produce important data to assess the real scope of the problem in each jurisdiction. This would then help gauge the need for any subsequent action at the national or European level. The research has found it very difficult to identify data to assess the exact scope of the problem. If no legislative measure is taken at the European level, a minimum option would ensure that a quantitative scope of the problem is known and would inform any future decision.

Such non-legislative measures should not duplicate existing actions already undertaken at national level.

Some stakeholders (defence lawyers, NGOs, academics) were of the view that non-legislative action would be insufficient; while few survey respondents shared their view on the case for EU intervention and what form it should take, the non-legislative option did not prove popular. Other stakeholders (still amongst lawyers, NGOs and academics) felt the extent to which non-legislative action could on its own achieve common minimum standards throughout the EU must be considered doubtful. Finally, some stakeholders (judges, but also lawyers and academics, as well as representatives of Ministries of Justice) believed that past experience showed that non-legislative action could be useful in some instances.

Policy option 3(a):

Option 3(a) involves the codification of ECtHR principles and on the top of that it provides for specific effective remedies to any breach of these principles. The fact that ECtHR principles are not codified in any comprehensive legislative instrument and the fact that those principles do not provide for prescription of any specific remedies in case of breach can be seen as a weakness of the current system. At least for certain aspects of presumption of innocence, introducing appropriate remedies in case of breach are justified. For reasons of coherence, under this option remedies are foreseen in case of breach of any of the aspects of presumption of innocence, and it will be afterwards assessed (in section 7) if such remedies are indeed justified in all cases.

The following example shows how the situation would improve under option 3(a): the suspect's or accused's right to remain silent would be established by option 3(a); if a person's right to remain silent is breached, currently in some Member States the only remedy is to raise this fact in a procedure of appeal. The specific effective remedy proposed under option 3(a) is a removal of evidence obtained in the breach of the right to remain silent from the court's procedure. This would provide a direct and immediate response to such breach. Currently some Member States have this remedy in place, however 12 Member States do not provide for any specific remedy during the proceedings of first instance and the only remedy is the possibility to appeal.

As regards opinion of stakeholders on legislative action, the Member States have already expressed a view that a legislative action on presumption of innocence might be needed in the framework of the Stockholm programme, in which they asked the Commission to examine the subject in detail. Some Member States (DE, FR, IT, SI) stated in the expert meeting they were in favour of legislative option, if it proves necessary. The majority of stakeholders, who filled in the on-line survey (defence lawyers, bar associations and academics) are in favour of legislative action taken by the EU as they believe the right to be presumed innocent is not sufficiently protected (see last diagram of Annex III).  

Policy option 3(b):

This option would involve the same as option 3(a) and, moreover, a definition of higher standards of protection than those included in option 3(a). The exception is the absence of public references to guilt, for which it would be irrealistic to establish a higher standard than the ECtHR standard and for which, therefore, options 3(a) and 3(b) coincide.

For the other aspects this option envisages higher standards of protection:

· Burden of proof and right of the accused to benefit from any doubt: in addition to the standards defined under option 3(a), exceptions to the general principle would be further limited by means of more strict proportionality considerations; such exceptions (where the burden of proof may be shifted to the defence) would be limited to minor offences and taking into account the seriousness of the sanction.

· Right not to incriminate oneself, right not co-operate and right to remain silent: higher standards than those under option 3(a) (and in the principles derived from the ECtHR case-law) would be defined. First, adverse inferences to be drawn from the right to be silent and the right not to cooperate would not be admitted; second, no exceptions would be admitted from the right not to cooperate (such as blood samples, bodily tissue for DNA testing), which would become absolute.

· Right to be present at one’s trial: again, the standards of protection under option 3(a) would be increased in order to ensure no exception, or at least more limited exceptions, to the rule that the accused has to be present at trial, so that even fewer trials take place without his presence.

As regards the opinion of stakeholders, Member States generally believed that this option is rather far-reaching. Some stakeholders (especially defence lawyers and NGOs) supported some of the aspects of option 3(b) (in particular the absolute right to remain silent).

Policy options 2, 3(a) and 3(b) are described in more detail in the table hereafter:

Objective || Policy option 2 [non-legislative measures] || Policy option 3(a) [ECtHR + remedies] || Policy option 3(b) [ECtHR plus + remedies]

Absence of public references to guilt before conviction || Draw up guidelines on the way in which these rights should be understood Set up a system to exchange best practice between Member States with a similar legal tradition Encourage the training of law enforcement officers and judicial authorities staff Encourage Member States to set up a monitoring system to collect relevant data || Establish the general principle: obligation for judicial authorities of not referring in public to a suspect or accused person as being guilty, before a final decision of a Court. + Specific remedies (request to remove the judicial authority concerned, financial compensation, re-trial) || No need for this option, since option 3(a) is sufficient.

Burden of proof and right of the accused to benefit from any doubt || Establish general principle that the burden of proof is on the prosecution and any doubt should benefit the accused; Allow for some exception (i.e. reversed burden of proof) under certain conditions: i) such presumption must always be rebuttable (i.e. there should always be a possibility to challenge a presumption of guilt); ii) proportionality principle (these presumptions must be confined within reasonable limits and maintain the rights of defence); + Specific remedies in case of breach (possibility of a re-trial – instead of simple right to appeal) || Option 3(a), but further clarifying the exceptions to the general principle by defining the proportionality principle (limiting the exceptions to minor offences and taking into account the seriousness of the sanction).

Right not to incriminate oneself, right not to cooperate and right to remain silent || Establish the general principle of the right not to incriminate one-self, the right not to cooperate and the right to silence (not allowing improper pressure from police or judicial authorities). Clarify circumstances under which adverse inferences can be drawn from the exercise of these rights according to the ECtHR case law. Allow for an exception for the right not to cooperate according to the ECtHR case law (blood samples, bodily tissue for DNA testing). + Specific remedies: minimum rules on the consequences of breach of these rights (declare the evidence obtained in breach of this right inadmissible in the court proceedings; establishing a right to re-trial rather than simple appeal or establishing a possibility to nullify the judgement). || Option 3(a) but not allowing adverse inferences to be drawn from the right not to cooperate and the right to be silent   No exceptions from the right not to cooperate.

Right to be present at one’s trial || Establish the general principles according to the ECtHR case law (a person charged with a criminal offence is entitled to take part in the hearing, and can be tried without being present only if he had an opportunity to waive his right to be present). + Specific remedy: right to re-trial (not only the right to appeal). || Option 3(a), but increasing the safeguards to ensure that even fewer trials take place without the presence of the accused (either according to CY and IE system, where no exceptions to the right to be present at trial are admitted or inspired by DE system, where there are only very limited exceptions to the right to be present: e.g. a person left in the middle of main hearing, de minimis rule, in certain appeal proceedings).

7.           Impact analysis of policy options

The impact analysis relies on in-depth analysis of the respective national legislation in EU Member States in order to assess the consequences of each parameter of each option.

As will be described hereafter, options 2, 3(a) and 3(b) are expected to have a positive impact on the fundamental rights of suspects or accused to be presumed innocent and to have access to a fair trial and to fully exercise the rights of defence.

All financial costs of each option are developed more in detail in Annex VI, which sets out the assumptions and calculations underlying the estimated expected impact of each option on each individual Member State. Exact data for calculation of the financial costs of the policy options were difficult to obtain. Only anecdotal evidence was available together with consultation of practitioners. These have shown that number of breaches can be counted in thousands rather than hundreds. Consequently, throughout the appendix, three scenarios have been used to calculate potential financial impacts of the different policy options. As regards estimated financial benefits in a form of reduction of current costs, particularly in the long term, this is based on the assumption that presumption of innocence will be more respected in the future as there will be clear binding rules at the EU level. Protection of these rules will be ensured by ECJ (preliminary rulings at a disposal of national courts and infringement procedure).

7.1         Policy option 1 - Status quo

Expected Impact

Effectiveness in achieving policy objective || · No incentive for any action to achieve better practical protection of the right to be presumed innocent and consequently no improvement of mutual trust in cross-border criminal proceedings. · Likely that no national reforms would take place at Member States’ initiative, only exceptionally as ad hoc responses to ECtHR case law · Insufficient standards and protection of the right to be presumed innocent in the EU would remain the same.

Social impact and Fundamental Rights || · No improvement in protection of the fundamental right to be presumed innocent. · Potential damage to: (i) suspects and accused persons’ fair trial rights; and (ii) the practical implementation of article 48 of the Charter.

Impact on the legal system of Member States || · Member States tend to interpret ECtHR pronouncements in different ways. Divergences between Member States' systems would remain as they continue to evolve along strictly national lines.

Financial and economic impact || · There are no immediate new financial burdens associated with this option. However, this option will not lead to the reduction of costs of ECtHR and domestic appeals, re-trials and financial compensation due to breach of suspects’ fair trial rights resulting from a failure of the right to be presumed innocent. In the absence of reliable data from Member States, current costs are tentatively assessed in Annex VI.

7.2         Policy option 2 - Non-legislative action

Expected Impact

Effectiveness in achieving policy objective || · Non-statutory initiatives (e.g. to: collect EU-wide data on breaches to the presumption of innocence; establish training programmes and non-statutory guidance for staff in competent authorities; share best practices in Member States) would all be of practical benefit. However, on its own, in the absence of a new legislative instrument, this option would be unlikely to ensure that Member States address existing problems related to the protection of the right to be presumed innocent or practical problems linked to its implementation. Consequently, no improvement of mutual trust in cross-border criminal proceedings is expected.

Social impact and Fundamental Rights || · Some improvement in the practical application of the right to be presumed innocent and thus suspects’ fair trial rights would be likely to accrue. But the absence of any method of enforcement means that implemented on its own there is likely to be only variable improvement between Member States.

Impact on the legal system of Member States || · Limited since the non-binding nature of this option would not on its own achieve common minimum standards throughout the EU. The guidelines, trainings may help the judiciary to interpret domestic provisions in compliance with the ECHR, but it is unlikely that this effect would be any more significant than the effect of ECtHR rulings alone. The Member States, which currently lack specific remedies in case of breach of presumption of innocence, are unlikely to introduce them in their legal system. · Proportionality of EU intervention would therefore not be at stake under this option and the question would rather be if it would suffice to fulfil the policy objectives.

Financial and economic impact || · There would be some limited set-up and running costs relating to the drafting of guidelines, organisation of the workshops, training and sharing of best practices. The financial burden on the Member States is estimated to be below 8 million euro per each of the four aspects of presumption of innocence referred to in option 3. Most of these costs (5.5 million euros) would relate to training costs for defence lawyers, police and judicial officers. However, if this option is put in place for all four issues together, the total costs are likely to be much lower than the total of the costs for each issue as there would be a high level of synergies in the training, workshops and staff for each[88]. · Establishing and operating a monitoring system in member States in view of fulfilling reporting obligations and collecting relevant data would entail an estimated cost of 1.3 million euros. · The estimated costs are per annum, except as regards drafting guidelines, which has an estimated one off cost of 47.520 euros. · The financial burden resulting from this option depends on the level of Member States' implementation of the guidelines.

7.3         Policy option 3(a) - Legislative action

Expected Impact

Effectiveness in achieving policy objective || · Because national judicial authorities would have greater mutual trust owing to the existence of common minimum standards for the provision of the right to be presumed innocent in all Member States, there would likely be fewer refusals to cooperate and fewer delays in cooperation with each other, with a corresponding diminution of the costs of associated delays, aborted proceedings, re-trials and appeals. · This option would effectively provide comprehensive provisions covering existing body of law and jurisprudence. In addition it would provide effective remedy to breach of the right to be presumed innocent. · Legislative option is enforceable contrary to non-legislative or status quo options. · Suspects or accused persons would benefit from common minimum standards in the area of presumption of innocence contained in Article 6(2) of the ECHR and further elaborated in ECtHR case law. In addition to this they would benefit from appropriate and effective remedies established in case of breach of these minimum standards. This will prevent the Member States from repeated violation of the ECtHR standards. · It would lead to less cases of miscarriages of justice, and there would not only be an improvement of the general perception of justice by suspects and accused, by victims, by judicial authorities, by defence lawyers, by the general public, but there would also be a reduction in the current costs for EU Member States resulting from a lack of or inadequate provision of the right to be presumed innocent. · Suspects would have a more effective redress mechanism against Member States in breach of Article 6(2) of the ECHR/Article 48 of the Charter via the ECJ than the ex-post complaint procedure of the ECtHR, where there is a significant backlog of cases and the court is unable to require infringing Member States to amend their national laws.

Social impact and Fundamental Rights || · Potential benefits to the fundamental rights of suspects and accused persons, and increased clarification of Article 48 of the Charter. · Risk of codifying the sometimes rather vague ECtHR jurisprudence through a binding legislative EU instrument. If case law develops towards a stronger protection in the future, a binding Directive establishing the present level of protection would not be up to date.

Impact on the legal system of Member States || · All Member States have enshrined the principle of presumption of innocence and related rights in their legislation and they already seem to be compliant with the ECtHR principles, therefore only minor legislative changes would be needed. · Proportionality of EU intervention would in a general way be respected, as the legislative changes required would be limited to some Member States having to put in place appropriate specific remedies in cases of breach of the rights already established in their legislation, or having to create a link between specific existing remedies and breach of this right. However, for some points of this option proportionality could be at stake. Mainly the practical implementation and application of these rights will have to be ensured by the Member States. · Justified doubts on the grounds of proportionality could be raised as regards specific remedies for the breach of absence of public references to guilt and burden and standard of proof. For these two aspects the analysis carried out has shown that the problem is not in the existing legislation and its remedies but rather on the lack of respect in practice; the level of intrusion of imposing Member States to lay down new specific remedies does not seem justified, as it is unlikely that the situation would change significantly. · As regards the right not to incriminate oneself, the right not to co-operate and the right to silence, some Member States would need to adapt their current remedies to the breach of this right in order to render these remedies more efficient. If the additional remedy is re-trial (instead of simple right to appeal), it would need to be introduced in all Member States except AT, FI, FR, HU. This might raise justified doubts in terms of proportionality, given that the consequence of a retrial is a fresh determination of the (whole) case, whereas the real question here is the legality of specific evidence. If the additional remedy is non-admissibility to the court of evidence obtained in breach of the right to remain silent, BE, BG, CY, EE, ES, HR, IE, LT, LV, NL, PL and SE would need to change their legislation. This aspect might raise a proportionality question from some Member States which could argue that the internal consistency of their judicial system could be at stake. Such arguments are difficult to accept: the principle of free evaluation of evidence should nevertheless allow excluding from the case evidence obtained in violation of fundamental rights, which seems to be in terms of legal certainty the correct means to ensure that the judge is not influenced by such evidence when taking the final decision. · As regards right to be tried in one´s presence, the following Member States would need to change their current legislation: BE, BG, HU and LV in order to allow for re-trial in case of breach of the right. Proportionality does not seem to be an issue in this point, in that what would be proposed is the minimum standards established in ECtHR case law.

Financial and economic impact || · Reduction in current costs of ECtHR and domestic appeals, re-trials, financial compensation, aborted prosecutions due to breach of suspects’ fair trial rights resulting from a lack of or inadequate provision of the right to be presumed innocent[89]. In particular in the long term, the financial impact estimated below should gradually reduce as the right to be presumed innocent should be more respected, and thus remedies for its breach would be less used. · Estimated financial impacts per annum for each aspect described below.

Absence of public references to guilt || · As all Member States already have this right established in the legislation as well as some remedies (possibility to remove the judge and pay damages), there would be no substantial costs in introducing this option. At the beginning there might be additional costs for Member States resulting from more persons asking for financial compensation for the breach of this right as they would be more aware of the possibility under a newly adopted Directive. However, in the long term, public authorities would learn to comply with this right and thus there would be savings of costs of financial compensation for breach of this right compared to status quo. · Costs for a possible extra remedy – re-trial (which would be introduced in all Member States except AT, FI, LT, PL and SE) are estimated to be 240.000 euros.

Burden of proof right of the accused to benefit from any doubt || · Costs for all Member States altogether are estimated between 92.000 and 920.000 euros. These are the costs of additional remedy (re-trial instead of simple right to appeal), which would need to be introduced in the Member States, which currently do not have it (all Member States except AT, FR and UK).

Right not to incriminate oneself, right not to co-operate and right to remain silent || · Costs for all Member States depend on the appropriate remedy established at the EU level, it would be for each Member State to choose, which of the following remedies is appropriate for their legal system. · If the additional remedy is re-trial (instead of simple right to appeal), which would need to be introduced in all Member States except AT, FI, FR, HU, the altogether costs are estimated between 98.000 and 980.000 euros. · If the additional remedy is non-admissibility to the court of evidence obtained in breach of the right to remain silent, costs would be incurred through the increase of prosecution activity in those Member States, where this remedy currently does not exist (BE; BG; CY, EE, ES, HR, IE, LT, LV, NL, PL, SE), (altogether estimated between 7.500 and 75.000 euros).

Right to be tried in one's presence || · Costs for all Member States altogether are estimated to 523.000 euros. These are the costs of additional remedy (re-trial instead of simple right to appeal), which would need to be introduced in the following Member States: BE, BG, HU and LV.

7.4         Policy option 3(b) - Legislative option

Expected Impact

Effectiveness in achieving policy objective || · Same as for option 3(a). · Mutual trust between judicial authorities would be further considerably strengthened as the level of minimum standards would be higher than in the present situation. This should lead to better judicial cooperation and fewer refusals and delays in mutual recognition of European Arrest Warrants and other judicial decisions. · In addition, suspects or accused persons benefit from higher common minimum standards in the area of presumption of innocence than those contained in Article 6(2) of the ECHR.

Social impact and Fundamental Rights || · The standards of protection of presumption of innocence would be even higher than in option 3a and thus fundamental rights would be better protected and no risk of non-compliance with future ECtHR jurisprudence should arise. · Potential benefits to the fundamental rights of suspects and accused persons, and increased clarification of Article 48 of the Charter. · Gradual culture change in the prosecution and judicial authorities on the respect of the right to be presumed innocent. · Adverse effect on administration of justice could occur as the rights of individuals would be strengthened to an extent which could perhaps harm efficiency of investigation and prosecution, misuse of justice could become more frequent.

Impact on the legal system of Member States || · All Member States have enshrined the principle of presumption of innocence and related rights in their legislation. However, the EU proposal would go further than the ECHR case law (setting higher standards to certain aspects of presumption of innocence); therefore several legislative changes would be necessary in certain Member States' legislations. The practical implementation and application of these rights will have to be ensured by the Member States · Proportionality of EU intervention might be questioned for some points of this option, in particular as regards the burden of proof and the right to be tried at one’s presence, where the higher standards proposed might have an important impact on prosecution activity. This impact might not be justified by the aim of ensuring high procedural rights standards, given that the ECtHR standards are in these points generally accepted as striking a correct balance between the public interest of crime investigation and prosecution and the rights of defence. · Conversely, the higher standard proposed for the right not to cooperate and the right to silence (not allowing inferences to be drawn from the exercise of such rights) does not seem to raise proportionality concerns, even if it is unlikely to be easily accepted by some Member States. This higher standard strikes a correct balance between the interests at stake: it is intended to protect the essence of the right to remain silent, so that a conviction should not be based on the silence of the accused and other incriminatory evidence should exist. · As regards burden of proof, some Member States will need to change their current legislation in order to limit the reversal of burden of proof: BE, HR, FR, HU, IE, PT, ES, SE and UK. · As regards right to remain silent, the following Member States would need to change their current legislation in order not to allow the possibility to draw adverse inferences from exercising this right: BE, CY, UK, FI, FR, IE, LV, NL and SE. · As regards right to be tried in one´s presence, all Member States would need to change their current legislation except CY, IE and DE.

Financial and economic impact || · Reduction in current costs of ECtHR and domestic appeals, re-trials, financial compensation, aborted prosecutions due to breach of suspects’ fair trial rights resulting from a lack of or inadequate provision of the right to be presumed innocent. In particular in the long term, the financial impact estimated below should gradually reduce as the right to be presumed innocent should be more respected, and thus remedies for its breach would be less used. · Estimated financial impacts per annum for each aspect described below.

Absence of public references to guilt || See option 3(a)

Burden of proof and right of the accused to benefit from any doubt || · This measure would increase prosecution activity in those Member States where the burden of proof can be currently reversed (there would be only limited exceptions to the reversal of burden of proof under option 3b) therefore additional cost might occur in BE, HR, FR, HU, IE, PT, ES, SE and UK. It is not possible to assess the exact financial impact; however the likely scenario estimates the costs to be altogether at 2,9 million euros.

Right not to incriminate oneself, right not to co-operate and right to remain silent || · This measure would increase prosecution activity in those Member States where the right to remain silent is not absolute (this system would be abolished under option 3(b) in BE, CY, UK, FI, FR, IE, LV, NL and SE). It is difficult to assess the exact financial impact; however, the costs in the most likely case scenario are tentatively estimated at 27 million euros altogether.

Right to be tried in one's presence || · Costs would arise for additional police resources used to ensure that a suspect or accused is physically brought to trial (currently tried in his absence under existing law) in all Member States except CY, IE and DE. Savings, on the other hand, would be generated by the avoided costs of all possible re-trials (if all persons are to be present at the trial no extra remedy would need to be used for breach of the right to be tried in one’s presence). It is not possible to know how many cases will be affected therefore the total cost is estimated at between 5.5 million euros and 22 million euros.

8.           Comparative assessment

Summary of possible policy options

Policy options || Effectiveness in achieving policy objective || Social Impact and Fundamental Rights || Financial and economic     impact || Impact on legal systems

Option 1: Status quo || X || X || X || X

Option 2: Non-legislative EU action || √ || √ || √ || √

Option 3 Legislative EU action : - (a) ECHR standards || √√ || √√ || √√ || √√

- (b) Beyond ECHR standards || √√√ || √√√ || √√√ || √√√

Key: X = little or no expected impact; √ = limited expected impact; √√ = high expected impact; √√√ very high expected impact.

If Policy Option 1 (status quo) is pursued the insufficient protection of the principle of presumption of innocence would continue and thus this option does not meet the identified objectives and is therefore not further considered. Generally, it is not very much supported by the stakeholders (defence lawyers, judges and academics). However, certain Member States are in favour of maintaining status quo.  

The measures envisaged within Policy Option 2 are likely to contribute to the objectives of an EU intervention in the field, but only to a limited extent. They are likely to have an impact on raising the awareness of stakeholders on the respect of presumption of innocence. However, Policy Option 2 is unlikely neither to strongly affect the application and enforcement of existing common standards, nor to improve the coherence of national legislations. In these circumstances mutual trust cannot be assured. Therefore Policy Option 2 is unlikely to sufficiently fulfil the objectives. Due to its limits this option did not prove to be very popular among stakeholders.

The legislative measures within Policy Option 3(a) and 3(b) are likely to contribute more effectively to the objectives of an EU intervention in the field. They would contribute (to varying degrees) to protection of the rights of suspects and accused persons to be presumed innocent by establishing minimum standards at the EU level for the protection of the presumption of innocence and appropriate remedies in case of breach. Some Member States were reluctant when discussing a possibility of legislative action taken by the EU; however, they had expressed their positive opinion in the way of asking the Commission to examine the subject of presumption of innocence in the Stockholm programme. The majority of other stakeholders (defence lawyers, NGOs, academics) are in favour of legislative action taken at the EU level (see last diagram of Annex III – analysis of the replies to the on-line survey).

Overall, Policy Option 3(b) is the most likely to meet all the objectives and to meet them to the greatest extent. However, it imposes additional obligations on Member States and foresees substantial legislative changes. As a result costs are likely to be considerably higher than in the other options, which is one of the reasons why it is the least feasible in terms of it being agreed. Most Member States and other stakeholders (except for some individual defence lawyers) are not in favour of option 3(b). Moreover, it would raise serious concerns in terms of subsidiarity and proportionality given the high intrusive legislative measures it involves, whereas in two of the specific problems (i.e., in the right not be pronounced guilty before conviction and in the burden of proof and standard of proof) the situation in Member States is satisfactory in legal terms and it is only the enforcement of the existing rules that has been revealed to be a problem. Indeed, less intrusive measures (such as the ones foreseen in Option 3(a) or, to a lesser extent, in Option 2) are likely to also meet the objectives.

In Policy Option 3(a) is likely to meet the objectives in a satisfactory way and the financial burden on the Member States is considerably lower than in Option 3(b).

In summary:

· the status quo option is unlikely to meet any of the operational objectives;

· the non-legislative option is only likely to meet the first two  operational objectives;

· the first legislative option (3(a) - ECtHR principles + remedies) is unlikely to meet all operational objectives – an adequate protection of the right not to incriminate one-self, the right not to co-operate and the right to remain silent is not guaranteed to be achieved under this option;

· the second legislative option (3(b) - beyond ECtHR principles + remedies) is likely to meet all the operational objectives. However, it is not needed as regards the first and second operational objectives (given that the first legislative option is enough).

9.           The preferred option

The preferred option is therefore a combination of elements from options 2, 3(a) and 3(b).

It can be summarised as follows: · Absence of public references to guilt before conviction - policy option 3(a), but without any specific remedy; · Burden of proof and any doubt on the guilt should benefit the accused - policy option 3(a), but without any specific remedy; · Right not to incriminate oneself, right not to co-operate and right to remain silent - combination of options 3(a) and 3(b): - Policy option 3(a), including specific remedies in case of breach; - Policy option 3(b) only as regards not allowing adverse inferences drawn from the exercise of these rights. · Right to be tried in one's presence – policy option 3(a), including specific remedies in case of breach; · Horizontal measures (policy option 2): - Training of the different actors involved in criminal proceedings; - Monitoring system on the situation of presumption of innocence in practice in the Member States.

The preferred option achieves all objectives set out in this impact assessment. As national judicial authorities would have greater mutual trust owing to the existence of common minimum standards for the provision of the right to be presumed innocent in all Member States, there would likely be fewer refusals to cooperate and fewer delays in cooperation with each other, with a corresponding diminution of the costs of associated delays, aborted proceedings, re-trials and appeals.

The preferred option fully respects the principles of subsidiarity and proportionality by proposing a differentiated EU-level intervention the definition of which was determined by several factors:

· Impact on the smooth functioning of mutual recognition instruments: particular attention should be given to those aspects which are closely linked and are indispensable to the proper functioning of mutual recognition instruments. Stronger intervention is justified where concrete procedural rights of suspect or accused persons – rather than general principle of procedural .criminal law – are at stake (such as for the right to silence and the right to be present at one’s trial);

· Level of protection by national law: stronger EU intervention is required for those aspects of presumption of innocence which are not adequately protected by national laws and where problems do not only lie with the practical application of these laws;

· Level of protection provided by ECHR: stronger EU intervention is required for those aspects where ECHR jurisprudence does not provide a standard which is sufficiently high in a common area of criminal justice (right to silence).

Absence of public references to guilt before conviction

For this aspect of presumption of innocence, the preferred option is policy option 3(a) but without any specific remedy. The explanation is:

· There exist problems with the practical enforcement of this aspect of presumption of innocence and it should therefore be covered by an EU legislative instrument given the added value of EU enforcement mechanisms.

· However, for reasons of proportionality and subsidiarity, given that all Member States have this aspect of presumption of innocence sufficiently protected in their legislation and also that the available remedies in case of breach seem as a whole satisfactory, it is not justified to introduce specific remedies.

· This aspect of presumption of innocence is only to a lesser extent linked to the functioning of the European Area of Justice, in the sense that it is a general principle of procedural criminal law -often a constitutional principle and, as such, less 'tangible'-, and is reflected in concrete procedural rights of suspected or accused persons in a rather indirect way.

· Given that no new remedy is to be proposed, no costs are foreseen for Member States. A reduction of existing costs is expected as regards those costs deriving from ECtHR and domestic appeals. In particular in the long term such costs will tend to reduce given that presumption of innocence would be more respected and thus remedies for each its breach would be less used.

Burden of proof and any doubt on the guilt should benefit the accused

For this aspect, the preferred option is also policy option 3(a) but again without any specific remedy. The explanation is similar to the one given for the previous aspect:

· There still exist problems with the practical enforcement of this aspect of presumption of innocence and it should therefore be covered by an EU legislative instrument given the added value of EU enforcement mechanisms.

· However, for reasons of proportionality and subsidiarity, given that all Member States have this aspect of presumption of innocence sufficiently protected in their legislation and also that the available remedies in case of breach seem as a whole satisfactory, it is not justified to introduce specific remedies.

· Again, this aspect of presumption of innocence is only to a lesser extent linked to the functioning of European Area of Justice, in the sense that it is a general principle of procedural criminal law -often a constitutional principle and, as such, less 'tangible'-, and is reflected in concrete procedural rights of suspect or accused persons in a rather indirect way.

· Given that no new remedy is to be proposed, no costs are foreseen for Member States. A reduction of existing costs is expected as regards those costs deriving from ECtHR and domestic appeals. In particular in the long term such costs will tend to reduce given that presumption of innocence would be more respected and thus there would be lesser remedies in case of breach.

Right not to incriminate oneself, right not to cooperate and right to remain silent

For this aspect, the preferred option is a combination of policy options 3(a) and 3(b).

· Establish the general principle of the right not to incriminate oneself, the right not to cooperate and the right to silence (not allowing improper pressure from police or judicial authorities) – option 3(a)-, but not allowing adverse inferences to be drawn from the exercise of those rights, thus going beyond the present ECHR standard –option 3(b).

· Allow for exceptions from the right not to cooperate according to the ECtHR case law (blood samples, bodily tissue for DNA testing –option 3(a)).

· Establish a specific remedy in case of breach: declare the evidence obtained in breach of these rights as inadmissible –option 3(a).

The explanation for this option is:

· Given the close link of this aspect of presumption of innocence to the functioning of European Area of Justice (e.g., with the EAW) - in the sense that it is establishes a concrete and 'tangible' procedural right of suspect or accused persons and not only a general principle of procedural criminal law, the general ECtHR principles need to be reinforced by not admitting inferences to be drawn from the exercise of these rights, without which their content is at risk and without which the objective of ensuring its protection is not ensured.

· To accompany this important change in several Member States, strong measures in case of breach are needed and a specific remedy in case of breach is justified, - non admissibility of evidence obtained in breach of these rights.

· The exceptions to the right not to cooperate identified by the ECtHR are justified by the fact that those are cases in which evidence has an existence independently of the will of the suspect or accused person, even though it might need to be obtained through the use of compulsory powers (see section 4.2.3).

· The estimated costs of this option are: (i) as regards establishing a prohibition to draw inferences from silence, the costs altogether for the 9 Member States which would need to change national legislation are tentatively estimated of 27 million euros in the most likely case scenario; (ii) as regards non admissibility of evidence obtained in breach of these rights, the estimated costs are altogether between 7.500 and 75.000 euros per annum for the 12 Member States (altogether) which would need to introduce this remedy in their national law.

Right to be tried in one's presence

For this aspect, the preferred option is policy option 3(a). The justification and explanation are:

· There exist problems with the practical enforcement of this aspect of presumption of innocence and it should therefore be covered by an EU legislative instrument given the added value of EU enforcement mechanisms.

· This aspect of presumption of innocence is also closely linked to the functioning of European Area of Justice (e.g., with the EAW), in the sense that it establishes a concrete and 'tangible' procedural right of suspect or accused persons, more than a general principles of procedural criminal law – for this reason, a specific remedy in case of breach is justified, also because 4 Member States do not have it – retrial.

· Such specific remedy is explicitly required by the ECtHR case law.

· The estimated costs of this option are 523.000 euros per annum for the 4 Member States (altogether) which currently do not have the possibility of a retrial.

Horizontal measures – training and monitoring (option 2)

· As described above, a system of training of different actors in criminal procedures (police, lawyers, prosecutors, judges etc) would support the implementaion of the legally-binding measures in several ways.  Training woul help legal officials to avoid discussing the suspect or accused and the relevant criminal proceedings in ways that could be seen to imply (or) the suspect’s guilt before a verdict had been reached. It would also ensure that legal officials are aware of thow to remedy breaches relating to unjustified reversal of the burden of proof. 

· A monitoring system is needed to produce more comprehensive and systematic data about the size and scope of the problem in each jurisdiction. This would help to assess the effectiveness of the intervention supported by this impact assessment, and to judge the need for any subsequent action at the national or European level.

The table below contains the total estimated costs of the preferred option per Member State and per aspect of presumption of innocence.

Costs of the preferred option for each Member State

EU jurisdiction/ Aspect of presumption of innocence || 1. Absence of public references to guilt No costs since no specific remedy || 2. Burden of proof and right to benefit from any doubt No costs since no specific remedy || 3. Right not to incriminate one-self, right not to cooperate and right to silence || 4. Right to be present at one's trial || Monitoring and evaluation for all four aspects[90] || Training for all four aspects[91] || Total Minimum / Maximum (for MS concerned)

Minimum cost of removal of evidence[92] (1/100,000 cases) || Maximum  cost of removal of evidence (1/10,000 cases) || Increased cost of prosecution[93] || Cost of retrial[94]

Austria || 0 || 0 || 0 || 0 || 0 || 0 ||  164.492 ||  140.672 || 305.164

Belgium || 0 || 0 ||  503 ||  5.035 ||  1.306.081 ||  257.870 ||  173.692 ||  109.084 ||  1.847.230 / 1.851.762  

Bulgaria || 0 || 0 ||  51 ||  515 || 0 || 0 ||  16.232 ||  37.298 ||  126.521 / 126.985

Croatia || 0 || 0 ||  143 ||  1.430 || 0 || 0 ||  58.192 ||  68.764 ||  127.099 / 128.386 

Cyprus || 0 || 0 ||  165 ||  1.647 ||  159.644 || 0 ||  101.004 ||  5.360 ||  266.173 / 267.655 

Czech Republic || 0 || 0 || 0 || 0 || 0 || 0 ||  76.320 ||  82.378 ||  158.698

Estonia || 0 || 0 || 0 || 0 || 0 || 0 ||  38.848 ||  17.824 ||  56.703 / 56.983

Finland || 0 || 0 || 0 || 0 ||  429.370 || 0 ||  108.756 ||  102.538 ||  640.664

France || 0 || 0 || 0 || 0 ||  7.757.873 || 0 ||  136.528 ||  888.752 ||  8.783.153

Germany || 0 || 0 || 0 || 0 || 0 || 0 ||  115.432 ||  6.514.856 ||  6.630.288

Greece || 0 || 0 || 0 || 0 || 0 || 0 ||  75.364 ||  115.426 ||  190.790

Hungary || 0 || 0 || 0 || 0 || 0 ||  168.680 ||  62.964 ||  207 ||  231.851

Ireland || 0 || 0 ||  173 ||  1.727 ||  438.540 || 0 ||  123.064 ||  28.824 ||  590.601 /  592.155 

Italy || 0 || 0 || 0 || 0 || 0 || 0 ||  95.400 ||  384.988 ||  480.388

Latvia || 0 || 0 || 0 || 0 ||  159.135 ||  24.000 ||  32.852 ||  66.010 ||  281.997

Lithuania || 0 || 0 ||  43 ||  433 || 0 || 0 ||  62.964 ||  30.650 ||  93.657 / 94.047

Luxembourg || 0 || 0 ||  49 ||  487 || 0 || 0 ||  197.264 ||  10.452 ||  207.765 / 208.203 

Malta || 0 || 0 || 0 || 0 || 0 || 0 ||  81.088 ||  1.762 ||  82.850

Netherlands || 0 || 0 ||  1.016 ||  10.161 ||  6.156.420 || 0 ||  179.240 ||  253.928 ||  6.590.604 / 6.599.749

Poland || 0 || 0 ||  981 ||  9.810 || 0 || 0 ||  61.056 ||  311.916 ||  373.953 / 382.782 

Portugal || 0 || 0 || 0 || 0 || 0 || 0 ||  73.456 ||  153.262 ||  226.718

Romania || 0 || 0 || 0 || 0 || 0 || 0 ||  21.680 ||  122.658 ||  144.338

Slovakia || 0 || 0 || 0 || 0 || 0 || 0 ||  42.764 ||  919.044 ||  961.808

Slovenia || 0 || 0 || 0 || 0 || 0 || 0 ||  80.136 ||  53.112 ||  133.248

Spain || 0 || 0 ||  2.701 ||  27.011 || 0 || 0 ||  97.612 ||  543.466 ||  643.779 / 668.089 

Sweden || 0 || 0 ||  202 ||  2.025 ||  1.273.164 || 0 ||  156.820 ||  159.434 ||  1.589.620 / 1.591.443

UK || 0 || 0 ||  1.564 ||  15.640 ||  9.347.852 || 0 ||  155.700 ||  159.434 ||  9.664.550 / 9.678.626 

NB:     The figures above have been rounded up from excel sheets.

10.         Transposition, monitoring and evaluation

With the preferred option the following mechanism will be used.

The timeframe for transposition of the Directive by Member States will be two years from its entry into force. As the Directive creates only a comparatively limited number of Member States' obligations (which, to some extent, mirror existing ECHR obligations or are obligations which already exist in a number of Member States), it is expected that a two-year deadline would provide Member States with sufficient time to effect necessary changes to their respective national laws and practice. The Commission will assist Member States and their national authorities in transposition of the Directive. Planned measures taken by the Commission aimed at countering any potentional difficulties of implementation will be listed in an Implementation Plan.

Providing for a robust monitoring and evaluation mechanism is crucial to ensure that the rights envisaged in the Directive are complied with in practice as well as in legislation. The Directive will stipulate that Member States should report on the effective implementation of legislative or non-legislative measures based on the nature of the proposed changes.

A legislative instrument (Directive) opens the possibility of EU enforcement mechanisms under Articles 258 and 259 of the TFEU and also of preliminary rulings under Article 267 of the TFEU. This is also an important element to be taken into account, as it ensures the effective transposition and implementation of the provisions laid down in that legislative instrument, if necessary with the intervention of the Court of Justice of the European Union. This represents an important step forward compared to the protection offered by the ECtHR, which does not have the power to ensure that its decisions are enforced.

Member States should be encouraged to collect relevant data to assist in this process as there is currently a lack of reliable empirical data.

The indicators that would be relevant to monitor the attainment of the objectives are summarised in the table below. 

Main policy objectives || Potential indicators for preferred option || Sources of information

To enhance mutual trust between Member States' authorities and thus facilitating mutual recognition of judgments and judicial decisions in the EU and improving judicial cooperation in the EU || · Number of refusals by Member State judicial authorities to execute another Member States' judicial decision on fair trial grounds, specifically on which component of fair trial · Member States' authorities' perception on the compliance with fair trial standards of proceedings in other Member States. This could be measured through questionnaires distributed to individual members of these authorities. || · Member States' governments, also: Eurojust · Judges' associations (e.g. European Network of Councils for the Judiciary)

To guarantee for EU citizens an effective high-level standard of protection of fundamental procedural rights in criminal procedure by: - ensuring that suspects and accused persons are presumed innocent and treated as such during the procedure - ensuring that authorities dealing with judicial cooperation and involved in the execution of a criminal sanction, of an investigation measure or of a European Arrest Warrant (EAW) are confident that the underlying decision was taken in full respect of the principle of presumption of innocence. || · Number of EAWs per MS (issuing executing) · Number of cases where EAWs are refused on PoI grounds · Number of cases which could be considered unfair or inadequate in terms of ECHR rights · Legislation passed by Member State to implement the changes by providing for the minimum common standards agreed · Number of successful and unsuccessful applications to the ECtHR where procedures that could be considered unfair or inadequate were used · Number of successful appeals against a final judgment or pre-trial decision on ground of procedures that could be considered unfair or inadequate || · Member States' governments · Member States' governments · Member States' governments · ECtHR · Member States' governments, Bar Associations

The Commission envisages carrying out specific empirical study with an emphasis on data collection 3-5 years into the implementation of the instrument[95], to gain in-depth quantitative and qualitative insights into the effectiveness of the proposal. All the data collected would enable the Commission to evaluate the actual compliance in Member States more robustly than using the means hitherto available.

ANNEX I Summary of cited ECtHR case law

Krause v Switzerland (Application 7986/77, judgement of 3 October 1978)

P. Krause was arrested in 1975 on suspicion of having committed various crimes in Switzerland and remanded in custody.  She was extradited to Italy in 1977.  Ahead of that, in 1976, the Federal Department of Justice and Police of Switzerland stated that P. Krause had “committed common law offences”.  The court, taking into consideration the whole interview stated that article 6(2) had not been violated.

Minelli v Switzerland (Application 8660/79, judgement of 25 March 1983)

The court held that the presumption of innocence would be violated if “without the accused’s having previously been proved guilty according to law and, notably, without his having had the opportunity of exercising his rights of defence, a judicial decision concerning him reflects an opinion that he is guilty. This may be so even in the absence of any formal findings; it suffices that there is some reasoning suggesting that the court regards the accused as guilty”.

Colozza v. Italy (Application 9024/80, judgment of 12 February 1985)

Mr Colozza was judged in Rome in his absence and sentenced to six years' imprisonment and a fine for various a offences, including fraud. However, certain services of the Rome public prosecutor’s office and of the Rome police had succeeded, in the context of other criminal proceedings, in obtaining Mr. Colozza’s address, which showed that it was thus possible to locate him. The Court concluded that Mr. Colozza had not waived exercise of his right to appear and to defend himself or that he was seeking to evade justice. The Court said that "When domestic law permits a trial to be held notwithstanding the absence of a person "charged with a criminal offence" (…), that person should, once he becomes aware of the proceedings, be able to obtain, from a court which has heard him, a fresh determination of the merits of the charge." It added that "the resources available under domestic law must be shown to be effective and a person "charged with a criminal offence" (…) must not be left with the burden of proving that he was not seeking to evade justice or that his absence was due to force majeure."

Such conditions were not met in Mr Colozza's case, and the Court therefore concluded that there was a breach of the requirements of Article 6(1) of the ECHR.

Salabiaku v. France (Application 10519/83, judgment of 7 October 1988)

This case concerned a person who had passed through customs with cannabis in a suitcase which he declared to be his property; he was convicted under the relevant provision of the Customs Code, which deems anyone carrying in contraband goods (consciously or not) guilty of an offence. Here the Strasbourg Court accepted the respondent state's argument that the strict liability offence was not disproportionate. The Court said that "Presumptions of fact or of law operate in every legal system. Clearly, the Convention does not prohibit such presumptions in principle. It does, however, require the Contracting States to remain within certain limits in this respect as regards criminal law". The Court added that "Article 6-2 does not therefore regard presumptions of fact or of law provided for in the criminal law with indifference. It requires States to confine them within reasonable limits which take into account the importance of what is at stake and maintain the rights of the defence."

Barberà, Messegué and Jabardo v. Spain (Application 10590/83, judgement of 6 December 1988)

The court held that the presumption of innocence requires, inter alia, that when carrying out their duties, the members of a court should not start with the preconceived idea that the accused has committed the offence charged; the burden of proof is on the prosecution, and any doubt should benefit the accused.

Funke v France (Application 10828/84, judgement of 25 February 1993)

Mr Funke started proceedings against France arguing that the country’s court’s demands that he produce documents amounted to an attempt to compel self-incrimination.  The Court agreed that the initial request for documents and the subsequent penalties were not unreasonable or contrary to the right of a fair trial - the request was a balanced part of a declaratory regime which saved individuals from strict and systematic investigation in return for their accepting certain duties and requirements; the subsequent penalties were a consequence of the refusal to cooperate.  However the Court decided that the customs authorities using the conviction of Funke in order to compel him to produce documents they believed to exist, without trying to procure the documents by other means, was a breach of Article 6-1.

Allenet de Ribemont v. France (Application no. 15175/89, judgement of 10 February 1995)

Following a press conference during which the director of the Paris criminal investigation department made the statement that “the haul was complete and the people involved in the case were under arrest”.  Mr de Ribemont took the matter to the ECtHR saying that France was in breach of articles 6(1) and 6(2) ECHR.  The court found that “the presumption of innocence [...] will be violated if a statement of a public official concerning a person charged with a criminal offence reflects an opinion that he is guilty before he has been proved guilty according to law. It suffices, even in the absence of any formal finding, that there is some reasoning suggesting that the court regards the accused as guilty “

John Murray v UK (Application 18731/91, judgement of 26 January 1996)

During his trial, Mr Murray, who had been arrest in Northern Ireland under the Prevention of Terrorism (Temporary Provisions) Act 1989, chose not to give evidence and to keep silent.  The court held that the right to remain silent was not absolute and that the drawing of reasonable inferences from silence was possible if, as was the case, it had not the effect of shifting the burden of proof from the prosecution to the defence so as to infringe the principle of the presumption of innocence.

Saunders v UK (Application 19187/91, judgement of 17 December 1996)

 In R v Saunders (1996) the accused was convicted on a number of counts of fraud relating to share dealing. During the investigation of the offence, the police relied on the Companies act (1985), which made it an offence to refuse to answer questions posed by fraud investigators. The ECtHR stated that "the public interest cannot be invoked to justify the use of answers compulsorily obtained in a non-judicial investigation to incriminate the accused during the trial proceedings" and "the prosecution in a criminal case [must] seek to prove their case against the accused without resort to evidence obtained through methods of coercion or oppression in defiance of the will of the accused."

Condron v. UK (Application 35718/97, judgement of 2 May 2000)

The applicants in this case decided to remain silent during police questioning following the advice of their solicitor, who was of the opinion that they were unfit to be interviewed due to the fact that they were suffering from heroin withdrawal symptoms. They were warned by the police that they did not have to say anything, but also that it might harm their defence if they did not mention when questioned something which they later rely on in court. The trials judge direction to the jury in that respect was in such terms that the ECtHR considered to have left the jury at liberty to draw an adverse inference from the accused's silence notwithstanding that it may have been satisfied as to the plausibility of the explanation. In the ECtHR's opinion, the jury should have been directed that it could only draw an adverse inference if satisfied that the applicants' silence at the police interview could only sensibly be attributed to their having no answer or none that would stand up to cross-examination. The ECtHR found there had been a violation of Article 6(1) of the ECHR and, as regards Article 6(2), considered that the applicants' argument amounted to a restatement of their case under Article 6(1); for that reason, it concluded that no separate issue arose under that head.

Daktaras v. Lithuania (Application 42095/98, judgement of 10 October 2000)

Mr Daktaras was portrayed in Lithuanian media as a local mafia chief. He complained that during his trial had not been impartial and that the prosecutor had commented that his guilt had been proved before the trial had started.  The court held that in asserting that the applicant's guilt had been "proved" by the evidence in the case-file, the prosecutor had used the same terminology as the applicant in his request to discontinue the case. The Court considered that, while the use of the term "proved" was unfortunate, both the applicant and the prosecutor were actually referring not to the question whether the applicant's guilt had been established by the evidence, but to the question whether the case-file had disclosed sufficient evidence of the applicant's guilt to justify proceeding to trial. The Court thus found no breach of article 6 paragraph 2.

Kudla v. Poland (Application 30210/96, judgment of 26 October 2000)

Mr Kudla was held in pre-trial detention for a period of a total of … months. The Court stated, first of all, that "the question of whether or not a period of detention is reasonable cannot be assessed in the abstract" and "must be assessed in each case according to its special features. Continued detention can be justified in a given case only if there are specific indications of a genuine requirement of public interest which, notwithstanding the presumption of innocence, outweighs the rule of respect for individual liberty laid down in Article 5 of the Convention". The Court added that it is for "National judicial authorities to ensure that, in a given case, the pre-trial detention of an accused person does not exceed a reasonable time."

The Court concluded that "The persistence of reasonable suspicion that the person arrested has committed an offence is a condition sine qua non for the lawfulness of the continued detention, but after a certain lapse of time it no longer suffices. The Court must then establish whether the other grounds given by the judicial authorities continued to justify the deprivation of liberty. Where such grounds were “relevant” and “sufficient”, the Court must also be satisfied that the national authorities displayed “special diligence” in the conduct of the proceedings (…)."

J.B. v. Switzerland (Application 31827/96, judgement of 5 March 2001)

The defendant alleged that he was obliged by the national courts to provide documents that could have incriminated him. The right not to incriminate oneself presupposes that the authorities seek to prove their case without resort to evidence obtained through methods of coercion or oppression in defiance of the will of the "person charged". By providing the accused with protection against improper compulsion by the authorities these immunities contribute to avoiding miscarriages of justice and securing the aims of article 6. The Court considered that there had been a violation of the right under article 6(1) not to incriminate oneself.

Telfner v. Austria (Application 33501/96, judgment of 20 March 2001)

The Court recalled its consistent jurisprudence according to which "Article 6 § 2 requires, inter alia, that when carrying out their duties, the members of a court should not start with the preconceived idea that the accused has committed the offence charged; the burden of proof is on the prosecution, and any doubt should benefit the accused (…). Thus, the presumption of innocence will be infringed where the burden of proof is shifted from the prosecution to the defence (…)."

Mr. Telfner was found guilty of a crime following a road accident in which his car had been involved. He did not reply to the police request to provide with information of who was the driver of the car at the moment when the accident had occurred and was convicted, in essence, "on a report of the local police station that the applicant was the main user of the car and had not been home on the night of the accident". The Court concluded that in requiring the applicant to provide an explanation although they had not been able to establish a convincing prima facie case against him, the courts shifted the burden of proof from the prosecution to the defence. Consequently, there was a violation of Article 6 § 2 of the Convention.

Butkevičius v. Lithuania (Application 48297/99, judgment of 26 March 2002)

Mr Butkevičius, who was Minister of Defence of Lithuania and a Member of the Parliament from 1996 to 2000, was arrested and referred to by the Prosecutor General and the Chairman of the Parliament, in the national press, with statements such as the following: “The Prosecutor General confirmed that [he had] enough sound evidence of the guilt of A. Butkevičius.”. "The Prosecutor General was quoted (…): 'I qualify the offence as an attempt to cheat… ". “When asked whether or not he doubts that A. Butkevičius accepted a bribe, the Chairman of the Seimas said: ‘on the basis of the material in my possession I entertain no doubt". “One or two facts were and are convincing. [The applicant] took the money while promising criminal services”.

The Court acknowledged that the fact that the applicant was an important political figure at the time of the alleged offence required the highest State officials, including the Prosecutor General and the Chairman of the Seimas, to keep the public informed of the alleged offence and the ensuing criminal proceedings. However, it cannot agree with the Government’s argument that this circumstance could justify any use of words chosen by the officials in their interviews with the press.

The Court also recalled that the presumption of innocence may be infringed not only by a judge or court but also by other public authorities. The Court concluded that there had been a breach of Article 6 § 2 of the Convention.

Kyprianou v. Cyprus (Application 73797/01, judgement of 15 December 2005)

The court held that “judicial authorities are required to exercise maximum discretion with regards to the cases with which they deal in order to preserve their image as impartial judges. [...]Thus, where a court president publicly used expressions which implied that he had already formed an unfavourable view of the applicant’s case before presiding over the court that had to decide it, his statements were such as to justify objectively the accused’s fears as to his impartiality”.

Jalloh v. Germany (Application 5481000, judgement of 11 July 2006)

Mr Jalloh was going arrested in Germany on the street under suspicion of being a drug dealer. When he was about to be arrested, he swallowed a little bad believed to contain drugs. The public prosecutor ordered that emetics be administered to Mr Jalloh by a doctor in order to provoke the regurgitation of the bag. He was taken to a hospital and, given that he refused to take the medication necessary to provoke vomiting, he was held down and immobilised by four police officers. The doctor then forcibly administered to him a salt solution and an emetic through a tube introduced into his stomach through the nose. In addition, the doctor injected him with apomorphine, another emetic that is a derivative of morphine. As a result, the applicant regurgitated one little bag containing 0.2182 grams of cocaine. Besides concluding that Mr Jalloh had been subjected to inhuman and degrading treatment contrary to Article 3 of the ECHR (on prohibition of torture), the ECtHR also held that allowing the use at Mr Jalloh's trial of evidence obtained by the forcible administration of emetics infringed his right not to incriminate himself and therefore rendered his trial as a whole unfair.

Pandy v. Belgium (Application 13583/02, judgment of 21 September 2006)

In the course of a criminal investigation into the applicant for several counts of murder, the investigating judge made remarks during a public hearing to the effect that the applicant should be comparing himself not with Dreyfus, but with Landru and Dr Petiot (two notorious serial killers). The applicant’s request for the judge to be withdrawn was dismissed. The appeal court found that the investigating judge had delivered an objective report on a difficult investigation and that the impugned remarks had been of minimal importance. The applicant was committed for trial before an assize court. The indictment was released to the press by a prosecution service spokesperson, as permitted by the law (the case was the subject of intense media interest), and was served on the applicant the same day, a few weeks ahead of the commencement of the proceedings before the assize court. The applicant was sentenced to life imprisonment for, among other offences, the murder of his two wives and four of his children and the rape and indecent assault of several of his daughters. The ECtHR found there had been a violation of presumption of innocence given that such statements involved a declaration of the accused's guilt which, firstly, encouraged the public to believe him guilty and, secondly, prejudged the assessment of the facts by the competent judicial authority.

Nestak v. Slovakia (Application 65559/01, judgement of 27 February 2007)

In a decision concerning pre-trial detention, the national courts stated that it had been proved that the applicant had committed the offence of which he had been charged. The ECtHR emphasised "that a fundamental distinction must be made between a statement that someone is merely suspected of having committed a crime and a clear judicial declaration, in the absence of a final conviction, that an individual has committed the crime in question". The Court concluded that "the statements impugned in the present case implied the applicant’s guilt before it was proved according to law". It added that "the fact that the applicant was ultimately found guilty and sentenced to a term of imprisonment cannot vacate his initial right to be presumed innocent until proved guilty according to law". There was accordingly a violation of Article 6 § 2 of the ECHR.

Tendam v. Spain (Application 25720/05, judgement of 13 July 2010)

Two sets of criminal proceedings were brought against the applicant. In the first set, he was detained pending trial for 135 days, and was subsequently convicted at first instance and acquitted on appeal. In the second set of proceedings he was likewise acquitted and sought the recovery of possessions seized from him during the investigation. Although some of the items were returned to him, he noticed that they were damaged and that others had disappeared. He unsuccessfully applied for compensation, both for the damage resulting from his pre-trial detention and for the malfunctioning of the justice system that had led to the failure to return the seized items or to their loss in value. His application was dismissed under both heads. In dismissing the applicant’s claim for compensation for his pre-trial detention, the Ministry of Justice and the Interior had relied on the fact that he had been acquitted on appeal for lack of sufficient evidence. Such reasoning, without qualification or reservation, cast doubt on the applicant’s innocence. In making a distinction between an acquittal for lack of evidence and an acquittal based on the finding that the alleged offence had not been committed, it had disregarded the applicant’s previous acquittal, which had to be taken into account by any judicial authority regardless of the reasons given for the criminal court’s decision. The national courts, for their part, had simply endorsed the Ministry’s reasoning without remedying the issue arising. The ECtHR therefore considered that there had been a breach of presumption of innocence.

Garlicki v. Poland (Application 36921/07, judgment of 14 June 2011)

Mr Garlicki, a doctor specialising in cardiac surgery, was arrested in the hospital where he was about to start a cardiac surgery by a dozen masked and armed officers of the Central Anti-Corruption Bureau (CAB). He was accused of medical negligence, harassment and receiving bribes from his patients. During a press conference, the head of the CAB referred to the defendant in the following terms: 'he is a ruthless and cynical bribe-taker. We have knowledge of several dozen bribes accepted by this doctor.' Furthermore, the Minister of justice made comments about the defendant that were deemed by the president of the constitutional court to have breached the constitution. The ECtHR recalled that this is a clear violation of presumption of innocence.

Poncelet v. Belgium (Application 44418/07, judgment of 30 March 2010)

The applicant was a senior civil servant. In 1994 an inspector was asked to conduct an administrative inquiry into certain public procurement contracts. He was of the opinion that there had been anomalies in the performance of those contracts and submitted various reports whose content displayed a hostile and biased attitude towards the applicant. In 1995 a judicial investigation was opened on charges of forgery and bribery. In 2006 the investigation division of the criminal court found that the inspector’s stance had breached the applicant’s right to be presumed innocent. In 2008 the criminal court, ruling on the merits after appeal proceedings before higher courts, came to the same conclusion. In 2009 the court of appeal declared the proceedings against the applicant admissible but found that the prosecution had become time-barred. It thus invalidated the effects of the investigation division’s decision and the criminal court’s judgment finding a breach of the right to be presumed innocent. The proceedings against the applicant having been brought and pursued in spite of the breach of the right to be presumed innocent and of defence rights, the court of appeal crystallised the feeling that only the limitation period had prevented the applicant’s conviction. The ECtHR therefore considered that there had been a breach of the applicant’s right to be presumed innocent.

Diacenco v. Romania

Mr Diacenco was involved in an accident in November 1998 in which the car he was driving hit a cart pulled by a horse, injuring one of the cart’s passengers. Indicted for battery, he was acquitted by the Court of Appeal in a final judgment of 9 July 2003. At the same time, that court held him criminally liable in the reasoning part of its judgment and ordered him to pay civil damages to the injured person. In fact, in seeking to protect the legitimate interests of the purported victim, the Court of Appeal expressly declared the applicant “guilty of the offence for which he was correctly indicted”. Consequently, the ECtHR considered that the language employed by the Court of Appeal overstepped the bounds of the civil forum, thereby casting doubt on the correctness of the acquittal. Accordingly, for the ECtHR there was a sufficient link to the criminal proceedings which was incompatible with the presumption of innocence.

Lagardère v. France (Application 18851/07, judgment of 12 April 2012)

In December 1992 a company lodged a complaint against Jean-Luc Lagardère, the applicant’s father, for misappropriation of corporate assets, and applied to join the criminal proceedings as a civil party. In June 1999 the father was brought before the criminal court, which declared the prosecution time-barred. In January 2002 the Paris Court of Appeal upheld all the provisions of that judgment. The company appealed on points of law. Jean-Luc Lagardère died in March 2003. In October 2003, after declaring that the prosecution had lapsed as a result of the accused’s death, the Court of Cassation quashed and annulled the civil provisions of the judgment of the Paris Court of Appeal and fixed a new, later date at which time had started to run for the purposes of the limitation period. The Versailles Court of Appeal, to which the case was referred for fresh examination, found that the constituent elements of the offence of misappropriation of corporate assets were established and ordered Jean-Luc Lagardère’s heirs to pay approximately fourteen million euros to the civil party. The applicant appealed on points of law, arguing that there had been a violation of Article 6 of the Convention because the criminal court had no authority to judge the matter after his father’s death. The Court of Cassation rejected the appeal.

The accused had died before his guilt had been lawfully established by a “tribunal”, so prior to his death he had been presumed innocent. Accordingly, in terms of both the language it had used and the reasoning it had given, the Versailles Court of Appeal had declared the applicant’s father guilty of the charges against him even though the prosecution had lapsed as a result of his death and no court had ever found him guilty during his lifetime. According to the EctHR, it had therefore violated his right to be presumed innocent.

ANNEX II List of stakeholders consulted - on-line survey on presumption of innocence

The stakeholders below were informed by the Commission about the on line survey carried out by CSES:

1) Members of the Expert Group on EU Criminal Policy

NAME || AFFILIATION || NATIONALITY

Petter Asp || Professor, University of Stockholm || SE

Luigi Foffani || Professor, University of Modena || IT

Dan Frände || Professor, University of Helsinki || FI

Estella Baker || Professor, University of Sheffield || UK

Berend Ferdinand Keulen || Professor, University of Groningen || NL

Valsamis Mitsilegas || Professor, University of London, Queen Mary || EL

Helmut Satzger || Professor, University of Munich || DE

Jocelyne Leblois-Happe || Professor, University of Strasbourg || FR

Pedro Caeiro || Professor, University of Coimbra || PT

Kristine Strada-Rozenberga || Professor, University of Latvia || LV

Jorge Espina || Prosecutor || ES

Joachim Ettenhofer || Prosecutor; EJN Contact Point || DE

Mike Kennedy || Crown Prosecution Service || UK

Galina Toneva || Deputy Prosecutor General || BG

Ignazio Patrone || Prosecutor General's Office || IT

Margarete von Galen || Defence Lawyer || DE

Paul Garlick || Barrister, QC || UK

Hans Sundberg || Judge, Court of Appeal || SE

Béatrice Blanc || Judge, President Tribunal de Grande Instance || FR

Igor Dzialuk || Independent Expert (former Undersecretary of State, Ministry of Justice and former Prosecutor) || PL

2) European Associations

1) International Association of Penal Law

2) Association of European Administrative Judges

3) Association of the Councils of State and Supreme Administrative Jurisdictions of the European Union

4) Council of Bars and Law Societies of European Commission

5) Council of the Notariats of the European Union

6) European Criminal Bar Association

7) ECLAN

8) European Association of Judges

9) European Judges and Prosecutors Association

10) European Judicial Network

11) European Union of Rechtspfleger

12) Ludwig Maximilians-University,

Institute for Criminal Law and Criminal Sciences

13) Justice

14) Network Of The Presidents Of The Supreme Judicial Courts Of The European Union

15) U.A.E. - Union des Avocats Européens

16) Victim support Europe

17) Amnesty International

18) Open Society

19) Fair Trial International

Annex III Overview of the responses to the survey questionnaire

ANNEX IV List of ECtHR cases in which EU Member States have been found in violation of presumption of innocence, between 1.1.2007 and 31.12.2012

Case || Application Number || Date of Judgement

GARYCKI v. POLAND || 14348/02 || 06/02/2007

NESTAK v. SLOVAKIA || 65559/01 || 27/02/2007

KAMPANELLIS v. GREECE || 9029/05 || 21/06/2007

VASSILIOS STAVROPOULOS v. GREECE || 35522/04 || 27/09/2007

SAMOILA AND CIONCA v. ROMANIA || 33065/03 || 04/03/2008

VITAN v. ROMANIA || 42084/02 || 25/03/2008

GEERINGS v. THE NETHERLANDS || 30810/03 || 01/03/2007

PARAPONIARIS v. GREECE || 42132/06 || 25/09/2008

RUPA v. ROMANIA || 58478/00 || 16/12/2008

NERATTINI v. GREECE || 43529/07 || 18/12/2008

DIDU v. ROMANIA || 34814/02 || 14/04/2009

PESA v. CROATIA || 40523/08 || 08/04/2010

TENDAM v. SPAIN || 25720/05 || 13/07/2010

PETYO PETKOV v. BULGARIA || 32130/03 || 07/01/2010

JIGA v. ROMANIA || 14352/04 || 16/03/2010

KRUMPHOLZ v. AUSTRIA || 13201/05 || 18/03/2010

FINSTER v. POLAND || 24860/08 || 08/02/2011

GIOSAKIS v. GREECE || 5689/08 || 03/05/2011

PONCELET v. BELGIUM || 44418/07 || 30/03/2010

TENDAM v. SPAIN || 25720/05 || 13/07/2010

KONSTAS v. GREECE || 53466/07 || 24/05/2011

LIZASO AZCONOBIETA v. SPAIN || 28834/08 || 28/06/2011

G.C.P. c. ROUMANIE || 20899/03 || 20/12/2011

PAVALACHE v. ROMANIA || 38746/03 || 18/10/2011

DIACENCO v. ROMANIA || 124/04 || 07/02/2012

LAGARDÈRE v. France || 18851/07 || 12/04/2012

ANNEX V Legal situation in the Member States regarding presumption of innocence ('PoI')

|| Absence of public references to guilt || Burden of proof and standard of proof || Right to remain silent, right not to cooperate and privilege against self-incrimination || In absentia decisions

How is this enshrined in law : Explicitly in constitution, criminal procedural code or case law - ○ Derived from the PoI ●[96] || Existence of specific remedies Yes – ○ No, but can be derived from general PoI - ● || The burden of proof always on the prosecution – ○ Reverse burdens of proof in certain circumstances - ● || Existence of specific remedies Yes – ○ No, but can be derived from general PoI - ● || Possibility to challenge the proof when reversed Yes – ○ No - ● || Right to remain silent is explicitly provided in law - ● Adverse inferences permitted - ◊ || Right not to cooperate and privilege against self-incrimination in national law - ○ The right is qualified: - where the information requested from the suspect exists independently of his will - ◊ - In certain additional circumstances ♦ No specific right not to cooperate and no privilege against self-incrimination-● Adverse inferences permitted - ■ || Existence of specific remedies Yes evidence can be inadmissible– ○ Yes decision can be nullified - ◊ No - ● || Explicit right to be present at trial - ● Suspect / accused’s presence at trial is without exceptions mandatory - ○ In absentia proceedings possible if: -Suspect / accused has voluntarily absconded - ◊ -In other circumstances - ♦ || If in absentia proceedings possible, the suspect / accused person may subsequently: -a fresh determination of the merits of the case - ○ -appeal the decision - ●

AT || ○ || ○ || ○ || ○ || N/A || ● || ○ ◊ || ◊ || ● ◊ || ○

BE || ○ || ● || ● || ● || ○ || ● ◊ || ● || ● || ● ◊ || ●

BG || ● || ● || ○ || ● || N/A || ● || ○ ◊ || ● || ● ◊ ♦ || ●

CY || ○ || ● || ○ || ● || N/A || ● ◊ || ○ ◊ || ● || ● ○ || N/A

CZ || ○ || ● || ○ || ● || N/A || ● || ○ ◊ || ○ || ● ◊ || ○

DE || ○ || ● || ○ || ● || N/A || ● ◊ || ○ ◊ ■ || ○ ◊ || ● ♦ || ○

DK || || || || || || || || || ||

EE || ○ || ● || ● || ● || ○ || ● || ● || ● || ● ◊ ♦ || ○

EL || ● || ● || ● || ● || ○ || ● || ● || ◊ || ● || ○

ES || ● || ● || ● || ● || ○ || ● || ○ ◊ || ● || ● ◊ || ○

FI || ○ || ○ || ○ || ○ || N/A || ● ◊ || ○ ◊ || ○ || ● ♦ || ○

FR || ○ || ● || ● || ○ || ○ || ● || ○ ◊ || ◊ || ● ◊ ♦ || ○

HU || ○ || ● || ● || ● || ○ || ● || ○ ◊ || ○ || ● ♦ || ●

HR || ○ || ● || ○ || ● || N/A || ● || ○ ◊ || ● || ● ◊ || ○

IE || ● || ● || ● || ● || ○ || ● ◊ || ○ ◊ ■ || ● || ○ || N/A

IT || || || || || || || || || ||

LT || ○ || ○ || ○ || ● || N/A || ● || ○ ◊ || ● || ● ♦ || ○

LU || ● || ● || ● || ● || ○ || ● || ○ ◊ || ○ || ● ◊ || ○

LV || ● || ● || ● || ● || ○ || ● ◊ || ○ || ● || ● || ●

MT || ● || ● || ● || ● || ○ || ● ◊ || ○ ◊ || ○ || ● ◊ || ○

NL || ● || ● || ○ || ● || N/A || ● ◊ || ● || ● || ♦ || ○

PL || ● || ○ || ● || ● || ○ || ● || ○ || ● || ● ◊ || ○

PT || ● || ● || ● || ● || ○ || ● || ● || ● || ● ◊ || ○

RO || ● || ● || || || || ● || ○ ◊ || ○ ◊ || ||

SE || ● || ○ || ● || ● || ○ || ● ◊ || ○ ◊ ■ || ● || ● ◊ ♦ || ○

SI || ● || ● || ○ || ● || N/A || ● || ○ ◊ || ○ || ● ◊ || ○

SK || || || || || || || || || ||

UK || E&W || || || ● || ○ || ○ || ◊♦ || ○ ◊ ♦[97] || ● || ● ◊ || ○

SC || ● || ● || ● || ○ || ○ || ◊♦ || ○ ◊ ♦ || ● || ● ◊ || ○

ANNEX VI Financial impacts

This Annex provides the detailed calculations used to assess the financial costs of the policy options

Nota Bene:

(1) Data of the extent of the problem have proved very difficult to obtain.  The research team has constantly sought to identify data on the number of breaches of each of these issues to no avail.  Ad hoc conversations with some practitioners and magistrates have subsequently shown that for each of the issues, the number of breaches per case can be counted in thousands rather than hundreds.  Consequently, throughout this appendix, three scenarios have been used to calculate to potential financial impacts of the different policy options. 

(2) The figures in the tables of this Annex have been rounded up from excel sheets and consequently do not always look to add up.

The right not to be referred to as guilty

Option 1 – Status Quo

Costs linked to damages

It is difficult to assess the current cost of the right not to be referred to as guilty as not data are available on the number of cases where the right not to be referred to as guilty is breached.  Only anecdotal evidence is available.  Any separate remedy of damages available to the “victim” would in most national legal systems potentially include both the restitution of pecuniary losses caused by the breach (as well as legal costs) and non-pecuniary damages to compensate for consequences such as distress or humiliation. In Garlicki –v- Poland where a doctor had been charged with corruption and homicide and his guilt had been alleged in slanderous terms by the Prosecutor General in a press conference, the victim brought a civil action against the Prosecutor General claiming PLN 70,000 non-pecuniary damages for loss of reputation, as well as demanding a public apology. The Polish court ordered an apology and awarded the doctor PLN 7,000 compensation for his moral suffering, and this sum was increased to PLN 30,000 (EUR 7,500) on appeal. In other cases, the ECtHR itself has considered the measure of damages for this type of breach and the awards it has made in this respect may not differ greatly from those provided by national courts[98]. By way of example, in Kyprianou –v- Cyprus[99] it awarded the victim EUR 15,000 as non-pecuniary damages for frustration and distress and in Pesa –v- Croatia[100] it awarded EUR 9,900 in a case of violation of the presumption of innocence.

Figures are also available for a number of ECtHR cases, namely Lizaso Azconobieta v. Spain, where the court ordered the responding country to pay damages of €12,000, G.C.P v. Romania where it was ordered to pay €2,000, Pavalach v. Romania (€10,000) and Kardas v. Greece (€12,000).

Based on an extrapolation using GDP per capita at PPP, the table below provides figures for each Member State of the damages that would be awarded in case of a breach of the right not to be referred to as guilty.

Column A – Member State

Column B – GDP per capita at PPP (base point)

Column C – Available data damages awarded per case (only one case known)

Column D – Amount of damage per base point (Column C / Column B)

Column E – Damages awarded per case (Column B x 135.43)

Table D1

A - EU jurisdiction || B - GDP/capita in PPS (2010) - EUROSTAT (base point) || C - Known damages || D - damages per base point (C/B) || E – Extrapolation – Cost of damages per case (B x average D)

Austria || 126 || || ||            17,064

Belgium || 119 || || ||            16,116

Bulgaria || 44 || || ||              5,959

Croatia || 61 || 9,900 ||        162.30 ||              8,261

Cyprus || 99 || 15,000 ||        151.52 ||            13,407

Czech Republic || 80 || || ||            10,834

Estonia || 64 || || ||              8,667

Finland || 115 || || ||            15,574

France || 108 || || ||            14,626

Germany || 118 || || ||            15,980

Greece || 90 || 12,000 ||        133.33 ||            12,188

Hungary || 65 || || ||              8,803

Ireland || 128 || || ||            17,335

Italy || 101 || || ||            13,678

Latvia || 51 || || ||              6,907

Lithuania || 57 || || ||              7,719

Luxembourg || 271 || || ||            36,701

Malta || 83 || || ||            11,240

Netherlands || 133 || || ||            18,012

Poland || 63 || 7,244 ||        114.98 ||              8,532

Portugal || 80 || || ||            10,834

Romania || 46 || 6,000 ||        130.43 ||              6,230

Slovakia || 74 || || ||            10,022

Slovenia || 85 || || ||            11,511

Spain || 100 || 12,000 ||        120.00 ||            13,543

Sweden || 123 || || ||            16,658

UK || 112 || || ||            15,168

Throughout the research phase, we have not been able to find any figures on the number of cases where the absence of public references to guilt had been breached.  Consequently, we have based the following extrapolations on different scenarios to provide some broad figures which would allow for some estimation of the costs of this option.  Ad hoc conversations with practitioners have indicated that cases where this right is breached are not common and were of the view that they are counted in breaches out of thousands rather than hundreds of cases.  We have thus developed three scenarios where 1/10,000, 1/50,000 and 1/100,000 criminal cases are awarded damages.

The table below provides estimates of the total cost of the status quo option based on the three scenarios.

Table D2

A - EU jurisdiction || Cost of damages per case || Number of criminal cases || 1 / 10,000 cases receiving damages || 1 / 50,000 cases receiving damages || 1 / 100,000 cases receiving damages

Austria ||   17,064 ||  60,726 ||  103,622 ||  20,724 ||   10,362

Belgium ||   16,116 ||   189,716 ||  305,744 ||  61,149 ||   30,574

Bulgaria ||  5,959 ||   118,262 || 70,470 ||  14,094 ||  7,047

Croatia ||  8,261 ||   110,524 || 91,304 ||  18,261 ||  9,130

Cyprus ||   13,407 ||   117,495 ||  157,529 ||  31,506 ||   15,753

Czech Republic ||   10,834 ||  97,675 ||  105,823 ||  21,165 ||   10,582

Estonia ||  8,667 ||  48,359 || 41,914 || 8,383 ||  4,191

Finland ||   15,574 ||  59,683 || 92,951 ||  18,590 ||  9,295

France ||   14,626 ||   1,061,097 ||  1,551,974 ||   310,395 || 155,197

Germany ||   15,980 ||   1,181,995 ||  1,888,875 ||   377,775 || 188,887

Greece ||   12,188 ||   195,929 ||  238,807 ||  47,761 ||   23,881

Hungary ||  8,803 ||   269,691 ||  237,403 ||  47,481 ||   23,740

Ireland ||   17,335 ||  77,625 ||  134,560 ||  26,912 ||   13,456

Italy ||   13,678 ||   1,607,646 ||  2,198,960 ||   439,792 || 219,896

Latvia ||  6,907 || 9,959 ||   6,878 || 1,376 ||  688

Lithuania ||  7,719 ||  81,277 || 62,741 ||  12,548 ||  6,274

Luxembourg ||   36,701 ||  14,579 || 53,506 ||  10,701 ||  5,351

Malta ||   11,240 ||  19,613 || 22,046 || 4,409 ||  2,205

Netherlands ||   18,012 ||   441,911 ||  795,961 ||   159,192 ||   79,596

Poland ||  8,532 ||   1,111,772 ||  948,553 ||   189,711 ||   94,855

Portugal ||   10,834 ||   115,466 ||  125,098 ||  25,020 ||   12,510

Romania ||  6,230 ||   171,480 ||  106,826 ||  21,365 ||   10,683

Slovakia ||   10,022 ||  41,189 || 41,278 || 8,256 ||  4,128

Slovenia ||   11,511 ||  90,205 ||  103,838 ||  20,768 ||   10,384

Spain ||   13,543 ||   1,336,505 ||  1,809,990 ||   361,998 || 180,999

Sweden ||   16,658 ||  91,000 ||  151,584 ||  30,317 ||   15,158

UK ||   15,168 ||   1,096,664 ||  1,663,402 ||   332,680 || 166,340

Total || || ||   11,752,184 ||   2,350,437 || 1,175,218

Based on those different scenarios, we can estimate the current situation (status quo) to have a financial impact of between and €1.2 million and €11.7 million per annum.

Due to the lack of precise data, we do not find it prudent to provide a set figure.

Furthermore, extra costs could be incurred by procedure being brought against a Member State, having the potential to be receivable by the ECtHR.  As an example, it is estimated that the Cadder case in Scotland cost in excess of €175,000[101].

Option 2 – Non-legislative option

Set up a system to exchange best practice to develop trust between lawyers, judicial and law enforcement authorities to develop understanding of each other’s role and reduce forcing suspects or accused persons to speak or cooperate. Draw up guidelines on the way in which this right should be understood.  Encourage the training of law enforcement officer and judicial authorities staff.

Under this option:

· Pan-European guidelines would be drawn up;

· Workshops would be organised to exchange best practices

· Monitoring and evaluation

· Training

Drawing up guidelines

The first step would be for the Commission to organise a series of 3 experts meetings. 

Based on the estimate that each EU workshop it would cost travel expenses for each MS expert at an average of €250, in addition to subsistence costs of €280, the total amount for the participants would be (€250+€280=) €530.  In addition, the cost of renting a room for the meeting as well as related food and drinks costs can be estimated at €1,000.  Overall, each workshop would cost a total of (€530 x 28 Member States =) €14,840 + €1,000 = €15,840.  The total annual cost of the workshops would thus be (€15,840 x 3 =) €47,520

The cost of disseminating the results can be seen as negligible as these can take the form of memos, internal guidelines or existing training could be amended to include the new guidelines. 

Best practice workshops

A series of workshops cold also be organised.  Each workshop would include

The impact of such workshops in individual Member States would depend on (i) what is discussed, (ii) the extent to which best practices are introduced in Member States with less efficient systems and (iii) other factors. 

In order to calculate the impacts of this option, we will assume that Member States will be divided into 6 groups (4 groups of 5 MS and 2 groups of 4 MS) with 5 workshops organised per group, making a total of 30 workshops. .

Assuming 60 people would participate in each workshop and based on the similar participations costs cost (travel expenses at an average of €250, in addition to subsistence costs of €280, the total amount for the participants would be (€250+€280=) €530), each workshop would cost €530 x 60 = € 31,800. 

Overall, the costs of organising the 30 workshops are as follow:

Travel + subsistence costs                                          €31,800                       x30      =         €954,000

Renting the room + logistics                                       €1,000             x30      =         €30,000

Total cost of Workshops                                                                                           €984,000

Monitoring and evaluation

In addition to the costs detailed above, the monitoring scheme would involve an estimated two full time employees either from the Ministry of Justice.

Based on the cost of the wage of an employee in the “business service” sector, the table below provides the costs of two full time equivalents (FTE) FTE in each of the MS. 

Table D3

|| Cost of FTE || Cost of 2 FTE

Belgium || 43,423 ||          86,846

Bulgaria || 4,058 ||            8,116

Czech Republic || 19,080 ||          38,160

Germany ||       28,858 ||          57,716

Estonia ||        9,712 ||          19,424

Ireland ||       30,766 ||          61,532

Greece ||       18,841 ||          37,682

Spain ||       24,403 ||          48,806

France ||       34,132 ||          68,264

Italy ||       23,850 ||          47,700

Cyprus ||       25,251 ||          50,502

Latvia ||        8,213 ||          16,426

Lithuania ||       15,741 ||          31,482

Luxembourg ||       49,316 ||          98,632

Hungary ||       15,741 ||          31,482

Malta ||       20,272 ||          40,544

Netherlands ||       44,810 ||          89,620

Austria ||       41,123 ||          82,246

Poland ||       15,264 ||          30,528

Portugal ||       18,364 ||          36,728

Romania ||        5,420 ||          10,840

Slovenia ||       20,034 ||          40,068

Slovakia ||       10,691 ||          21,382

Finland ||       27,189 ||          54,378

Sweden ||       39,205 ||          78,410

United Kingdom ||       38,925 ||          77,850

Croatia ||       14,548 ||          29,096

|| || 1,294,460

Source: Eurostat

Overall, monitoring and evaluation would thus cost €1.3 million

Training

Based on a 2007 study[102], the hourly fee of a lawyer in the EU is estimated to be between €51 and €574.  Assuming that the legal experts providing the training charge half the price of a lawyer, we estimated that the training session would be calculated assuming a length of 2 hours.  The cost of the trainer therefore ranges between €51 and €574.

Further assuming that the training would be done in groups of 10 staff (judge and prosecutors), and that all judges and prosecutors in Member States will receive the training, the costs are presented in the following table.  For instance, in Austria, the cost of the trainer is assumed to be €383 per session.  Considering there are 1,837 staff to train, there would be a total of 184 training session, for a total cost of €70,336.

In addition, we assume that the costs of the premises do not have to be taken into account as the training sessions will take place in rooms belonging to the judicial system (tribunals etc...). Finally, we have excluded travel costs as we assume that due to the limited number of participants in the training sessions, those could take place in different local premises, rather than in the capital city. 

The total cost of training would stand at € 5.5 million

Table D4

EU jurisdiction || Number of judges || Number of prosecutors || Cost of trainer || number of staff to train || number of training sessions || cost of training

Austria || 1,491 || 346 || 383 || 1,837 || 184 || 70,336

Belgium || 1,607 || 835 || 223 || 2,442 || 244 || 54,542

Bulgaria || 2,198 || 1,455 || 51 || 3,653 || 365 || 18,649

Croatia || 1,887 || 619 || 137 || 2,506 || 251 || 34,382

Cyprus || 104 || 106 || 128 || 210 || 21 || 2,680

Czech Republic || 3,063 || 1,240 || 96 ||  4,303 ||     430 ||   41,189

Estonia || 224 || 175 || 223 || 399 || 40 || 8,912

Finland || 967 || 372 || 383 || 1,339 || 134 || 51,268

France || 9,645 || 1,961 || 383 || 11,606 || 1,161 || 444,376

Germany || 79,832 || 5,244 || 383 || 85,076 || 8,508 || 3,257,428

Greece || 2,041 || 543 || 223 || 2,584 || 258 || 57,713

Hungary || 2,891 || 1,741 || 223 || 4,632 || 463 || 103,455

Ireland || 147 || 191 || 574 || 338 || 34 || 19,412

Italy || 6,654 || 1,978 || 223 || 8,632 || 863 || 192,494

Latvia || 472 || 390 || 383 || 862 || 86 || 33,005

Lithuania || 767 || 834 || 96 || 1,601 || 160 || 15,325

Luxembourg || 188 || 46 || 223 || 234 || 23 || 5,226

Malta || 39 || 30 || 128 || 69 || 7 || 881

Netherlands || 2,530 || 786 || 383 || 3,316 || 332 || 126,964

Poland || 10,625 || 5,668 || 96 || 16,293 || 1,629 || 155,958

Portugal || 1,956 || 1,475 || 223 || 3,431 || 343 || 76,631

Romania || 4,081 || 2,326 || 96 || 6,407 || 641 || 61,329

Slovakia || 1,351 || 935 || 96 || 2,286 || 229 || 21,882

Slovenia || 1,024 || 165 || 223 || 1,189 || 119 || 26,556

Spain || 4,689 || 2,408 || 383 || 7,097 || 710 || 271,733

Sweden || 1,081 || 1,001 || 383 || 2,082 || 208 || 79,717

UK || 1,984 || 3,035 || 574 || 5,019 || 502 || 288,255

Total || 143,538 ||        35,905 || || || || 5,520,298

Source: CEPEJ, CSES

Overall, if all the elements of this option are put in place, we can estimate the total cost to be:

Guidelines (one off cost)                                             €47,520

Workshops (annually)                                      €984,000

Evaluation and monitoring (annually)             €1,294,460

Training (annually)                                          €5,520,298

Total                                                                €7,846,278

Option 3 – Legislative option

Set out the obligation for judicial authorities and their representatives not to refer in public to a suspect or accused person as being guilty, before the final decision of a court.  This principle would be set out without any exceptions.  Furthermore, the option would set out appropriate remedies in case of breach of the principle

The cost of this option would be linked to the number of retrials as the remedy would have to take place before the final decision of a court.

Cost of retrial

No data exist for the cost of prosecution purely in criminal cases.  Based on the data available, it is only possible to calculate a cost of the prosecution per case in all judicial procedures.  Assuming that 65% of judicial cases in the EU are criminal cases[103], we can extrapolate the total number of judicial cases for each Member State.  By taking the total public budget allocated to all courts (CEPEJ), we can then estimate the cost of a procedure.  Based on the assumptions set out under option 1 that the number of breaches can be counted per thousand cases rather than hundred cases, we provide the same three scenarios (1/10,000, 1/50,000 and 1/100,000) that would require a re-trial, we can therefore estimate the extra procedural costs in each Member State.

Column A – Member State

Column B – Number of criminal cases (CEPEJ)

Column C – Total number of cases (civil + criminal) = Column B / 65 x 100

Column D – Total annual approved public budget – where this was not available (*), we have calculated the Total annual approved public budget per inhabitant and have extrapolated it by the population

Column E – Population

Column F – Total annual approved public budget per inhabitant (Column D / Column E)

Column G – Cost per case (Column D / Column F)

Column H – Number of cases leading to a retrial (Column B / scenario )

Column I – Cost of retrial = (Column G x Column H)

Table D5

A EU jurisdiction || B Number of criminal cases (CEPEJ) || C Total number of cases (civil and criminal) || D Total annual approved public budget (CEPEJ) || E Population || F Total annual approved public budget per inhabitant || G Cost per case || 1/10,000 || 1/50,000 || 1/100,000

H Number of cases leading to a retrial || I Cost of retrial || H Number of cases leading to a retrial || I Cost of retrial || H Number of cases leading to a retrial || I Cost of retrial

Austria ||        60,726 ||            93,425 || 304,454,631 || 8,404,252 || ||          3,259 ||                  6 ||             19,790 ||                  1 ||             3,958 ||                 1 ||            1,979

Belgium ||      189,716 ||          291,871 || 396,723,427 || 10,951,266 || ||          1,359 ||                19 ||             25,787 ||                  4 ||             5,157 ||                 2 ||            2,579

Bulgaria ||      118,262 ||          181,942 || 112,211,184 || 7,504,868 ||                15 ||             617 ||                12 ||               7,294 ||                  2 ||             1,459 ||                 1 ||                729

Croatia ||      110,524 ||          170,037 || 211,304,301 || 4,412,137 ||                48 ||          1,243 ||                11 ||             13,735 ||                  2 ||             2,747 ||                 1 ||            1,373

Cyprus ||      117,495 ||          180,762 || 33,546,827 || 804,435 ||                42 ||             186 ||                12 ||               2,181 ||                  2 ||                436 ||                 1 ||                218

Czech Republic ||        97,675 ||          150,269 || 346,497,809 || 10,532,770 ||                33 ||          2,306 ||                10 ||             22,522 ||                  2 ||             4,504 ||                 1 ||            2,252

Estonia ||        48,359 ||            74,398 || 26,797,340 || 1,340,194 ||                20 ||             360 ||                  5 ||               1,742 ||                  1 ||                348 ||                 0 ||                174

Finland ||        59,683 ||            91,820 || 243,066,350 || 5,375,276 ||                45 ||          2,647 ||                  6 ||             15,799 ||                  1 ||             3,160 ||                 1 ||            1,580

France ||   1,061,097 ||       1,632,457 || 2,356,460,787 || 65,048,412 || ||          1,444 ||              106 ||           153,170 ||                21 ||           30,634 ||               11 ||          15,317

Germany ||   1,181,995 ||       1,818,454 || 2,961,554,917 || 81,751,602 || ||          1,629 ||              118 ||           192,501 ||                24 ||           38,500 ||               12 ||          19,250

Greece ||      195,929 ||          301,429 || 409,714,852 || 11,309,885 || ||          1,359 ||                20 ||             26,631 ||                  4 ||             5,326 ||                 2 ||            2,663

Hungary ||      269,691 ||          414,909 || 259,501,133 || 9,985,722 ||                26 ||             625 ||                27 ||             16,868 ||                  5 ||             3,374 ||                 3 ||            1,687

Ireland ||        77,625 ||          119,423 || 148,722,000 || 4,480,858 ||                33 ||          1,245 ||                  8 ||               9,667 ||                  2 ||             1,933 ||                 1 ||                967

Italy ||   1,607,646 ||       2,473,302 || 3,051,375,987 || 60,626,442 ||                50 ||          1,234 ||              161 ||           198,339 ||                32 ||           39,668 ||               16 ||          19,834

Latvia ||          9,959 ||            15,322 || 36,919,820 || 2,229,641 ||                17 ||          2,410 ||                  1 ||               2,400 ||                  0 ||                480 ||                 0 ||                240

Lithuania ||        81,277 ||          125,042 || 50,567,945 || 3,244,601 ||                16 ||             404 ||                  8 ||               3,287 ||                  2 ||                657 ||                 1 ||                329

Luxembourg ||        14,579 ||            22,429 || 18,542,050 || 511,840 || ||             827 ||                  1 ||               1,205 ||                  0 ||                241 ||                 0 ||                121

Malta ||        19,613 ||            30,174 || 10,260,000 || 417,617 ||                25 ||             340 ||                  2 ||                  667 ||                  0 ||                133 ||                 0 ||                  67

Netherlands ||      441,911 ||          679,863 || 990,667,000 || 16,655,799 ||                59 ||          1,457 ||                44 ||             64,393 ||                  9 ||           12,879 ||                 4 ||            6,439

Poland ||   1,111,772 ||       1,710,418 || 1,365,085,000 || 38,200,037 ||                36 ||             798 ||              111 ||             88,731 ||                22 ||           17,746 ||               11 ||            8,873

Portugal ||      115,466 ||          177,640 || 528,943,165 || 10,636,979 ||                50 ||          2,978 ||                12 ||             34,381 ||                  2 ||             6,876 ||                 1 ||            3,438

Romania ||      171,480 ||          263,815 || 355,246,737 || 21,390,000 ||                17 ||          1,347 ||                17 ||             23,091 ||                  3 ||             4,618 ||                 2 ||            2,309

Slovakia ||        41,189 ||            63,368 || 138,493,788 || 5,435,273 ||                25 ||          2,186 ||                  4 ||               9,002 ||                  1 ||             1,800 ||                 0 ||                900

Slovenia ||        90,205 ||          138,777 || 178,158,919 || 2,050,189 ||                87 ||          1,284 ||                  9 ||             11,580 ||                  2 ||             2,316 ||                 1 ||            1,158

Spain ||   1,336,505 ||       2,056,162 || 1,671,947,969 || 46,152,926 || ||             813 ||              134 ||           108,677 ||                27 ||           21,735 ||               13 ||          10,868

Sweden ||        91,000 ||          140,000 || 557,260,358 || 9,415,570 ||                59 ||          3,980 ||                  9 ||             36,222 ||                  2 ||             7,244 ||                 1 ||            3,622

UK ||   1,096,664 ||       1,687,175 || 1,411,574,820 || 62,641,000 ||                23 ||             837 ||              110 ||             91,752 ||                22 ||           18,350 ||               11 ||            9,175

Total || || || || || || ||     982 ||    1,181,414 ||     196 ||       236,283 ||      98 ||     118,141

Overall, the total cost of this option would be between €120,000 and €1.2 million. Taking the median scenario, we can estimate the costs of this option to stand at €240,000 per annum.

Burden of proof and standards of proof

Option 1 – Status Quo

It is difficult to assess the current financial impacts linked to the burden and standards of proof. However, one can assume that the reversal of the burden of proof in some Member States lead to a reduction in the cost of prosecution which can be calculated as being equivalent to the cost of the introduction of Option 3(b): € 55 million.

Furthermore, extra costs could be incurred by procedure being brought against a Member State, having the potential to be receivable by the ECtHR.  As an example, it is estimated that the Cadder case in Scotland cost in excess of €175,000[104].

Option 2- Non-legislative option

Set up a system to exchange best practice to develop trust between lawyers, judicial and law enforcement authorities to develop understanding of each other’s role and reduce forcing suspects or accused persons to speak or cooperate. Draw up guidelines on the way in which this right should be understood.  Encourage the training of law enforcement officer and judicial authorities staff

The cost of Option 2 would be similar for each issue.  Consequently, the cost of this option can be estimated to reach €7,846,278.  If Option 2 is chosen for each of the research stands however, the total cost would be significantly lower than €7,846,278 x 4 = €31,382,112 as some significant synergies and saving would take place.

Option 3(a) – Legislative option (ECHR standards)

Establish general principle that the burden of proof is on the prosecution and the doubt should benefit the accused.

Allow for some exceptions (i.e., reversed burden of proof) under certain conditions:

i) such presumption must always be rebuttable (i.e. there should always be a possibility to challenge a presumption of guilt);

ii) proportionality principle (these presumptions must be confined within reasonable limits and maintain the rights of defence);

+ Remedies

Cost of retrial

The cost of this option would be linked to the remedies set out (cost of re-trial).  The table below provides the costs of re-trial based on the assumption that one in 10,000, 50,000 and 100,000 cases will lead to a retrial.

Table D6

EU jurisdiction || Cost per case || 1/10,000 || 1/50,000 || 1/100,000

Number of cases leading to a retrial || cost of retrial || Number of cases leading to a retrial || cost of retrial || Number of cases leading to a retrial || cost of retrial

Austria ||          3,259 ||    6 ||    19,790 ||    1 ||      3,958 ||        1 ||       1,979

Belgium ||          1,359 ||       19 ||    25,787 ||    4 ||      5,157 ||        2 ||       2,579

Bulgaria ||             617 ||       12 ||      7,294 ||    2 ||      1,459 ||        1 ||      729

Croatia ||          1,243 ||       11 ||    13,735 ||    2 ||      2,747 ||        1 ||       1,373

Cyprus ||             186 ||       12 ||      2,181 ||    2 ||    436 ||        1 ||      218

Czech Republic ||          2,306 ||       10 ||    22,522 ||    2 ||      4,504 ||        1 ||       2,252

Estonia ||             360 ||    5 ||      1,742 ||    1 ||    348 ||        0 ||      174

Finland ||          2,647 ||    6 ||    15,799 ||    1 ||      3,160 ||        1 ||       1,580

France ||          1,444 ||     106 ||       153,170 ||       21 ||    30,634 ||     11 ||     15,317

Germany ||          1,629 ||     118 ||       192,501 ||       24 ||    38,500 ||     12 ||     19,250

Greece ||          1,359 ||       20 ||    26,631 ||    4 ||      5,326 ||        2 ||       2,663

Hungary ||             625 ||       27 ||    16,868 ||    5 ||      3,374 ||        3 ||       1,687

Ireland ||          1,245 ||    8 ||      9,667 ||    2 ||      1,933 ||        1 ||      967

Italy ||          1,234 ||     161 ||       198,339 ||       32 ||    39,668 ||     16 ||     19,834

Latvia ||          2,410 ||    1 ||      2,400 ||    0 ||    480 ||        0 ||      240

Lithuania ||             404 ||    8 ||      3,287 ||    2 ||    657 ||        1 ||      329

Luxembourg ||             827 ||    1 ||      1,205 ||    0 ||    241 ||        0 ||      121

Malta ||             340 ||    2 ||    667 ||    0 ||    133 ||        0 ||        67

Netherlands ||          1,457 ||       44 ||    64,393 ||    9 ||    12,879 ||        4 ||       6,439

Poland ||             798 ||     111 ||    88,731 ||       22 ||    17,746 ||     11 ||       8,873

Portugal ||          2,978 ||       12 ||    34,381 ||    2 ||      6,876 ||        1 ||       3,438

Romania ||          1,347 ||       17 ||    23,091 ||    3 ||      4,618 ||        2 ||       2,309

Slovakia ||          2,186 ||    4 ||      9,002 ||    1 ||      1,800 ||        0 ||      900

Slovenia ||          1,284 ||    9 ||    11,580 ||    2 ||      2,316 ||        1 ||       1,158

Spain ||             813 ||     134 ||       108,677 ||       27 ||    21,735 ||     13 ||     10,868

Sweden ||          3,980 ||    9 ||    36,222 ||    2 ||      7,244 ||        1 ||       3,622

UK ||             837 ||     110 ||    91,752 ||       22 ||    18,350 ||     11 ||       9,175

Total (minus AT, FR, UK) || || 760 || 916,702 || 152 || 183,340 || 76 || 91,670

The total cost of this option would thus be between € 92,000 and €920,000 per annum.

Option 3 (b) – Legislative option (beyond ECHR standards)

Establish general principle that the burden of proof is on the prosecution and the doubt should benefit the accused.

Do not allow any exceptions (i.e., reversed burden of proof)

+ Remedies

This option would lead to an increase in the costs at the investigative stage, which can be translated into an increase in the cost of the prosecution.  This is calculated below.

Increased costs of the prosecution

The table below provides figures on the total annual budget allocated for public prosecution.  Figures are taken from the CEPEJ report (2012), providing 2010 data.  Figures followed by an asterisk (*) denote Member States for which the data was not provided.  We have calculated the Public annual budget allocated to public prosecution per inhabitant where possible and extrapolated the data based on the average (€11.9).  Overall, we estimate that the total annual budget allocated to public prosecution in the EU (with the exception of Denmark and the addition of Croatia) is €6,737,848,858 (€ 6.8 billion)

Table D7

EU jurisdiction || Public annual budget allocated to public prosecution || Population || Public annual budget allocated to public prosecution per inhabitant

Austria ||              100,231,686* ||          8,404,252 || n/a

Belgium ||              130,608,155* ||        10,951,266 || n/a

Bulgaria ||                79,203,203 ||          7,504,868 || 10.6

Croatia ||                41,296,176 ||          4,412,137 || 9.4

Cyprus ||                15,964,412 ||             804,435 || 19.8

Czech Republic ||                83,446,289 ||        10,532,770 || 7.9

Estonia ||                  9,135,614 ||          1,340,194 || 6.8

Finland ||                42,937,000 ||          5,375,276 || 8.0

France ||              775,787,302* ||        65,048,412 || n/a

Germany ||              974,994,666* ||        81,751,602 || n/a

Greece ||              134,885,155* ||        11,309,885 || n/a

Hungary ||              102,321,320 ||          9,985,722 || 10.2

Ireland ||                43,854,000 ||          4,480,858 || 9.8

Italy ||           1,249,053,619 ||        60,626,442 || 20.6

Latvia ||                15,913,545 ||          2,229,641 || 7.1

Lithuania ||                29,555,000 ||          3,244,601 || 9.1

Luxembourg ||                  6,104,361* ||             511,840 || n/a

Malta ||                  2,569,000 ||             417,617 || 6.2

Netherlands ||              615,642,000 ||        16,655,799 || 37.0

Poland ||              312,514,570 ||        38,200,037 || 8.2

Portugal ||              113,901,622 ||        10,636,979 || 10.7

Romania ||              162,428,333 ||        21,390,000 || 7.6

Slovakia ||                63,702,886 ||          5,435,273 || 11.7

Slovenia ||                19,263,376 ||          2,050,189 || 9.4

Spain ||              550,433,943* ||        46,152,926 || n/a

Sweden ||              127,316,425 ||          9,415,570 || 13.5

UK ||              934,785,200 ||        62,641,000 || 14.9

Total ||           6,737,848,858 || ||

Total MS where burden of proof can be reversed (BE, HR, FR, HU, IE, PT, ES, SE, UK) ||           2,820,304,143 || ||

While it is not possible to assess the exact financial impact of the increased prosecution costs, we can provide some likely scenarios.

In the case of Option 3(b), prosecution activity would increase in those MS where the burden of proof can currently be reversed (BE, HR, FR, HU, IE, PT, ES, SE, UK).  Below we provide different scenarios.  It is very difficult to assess the increase of prosecution activity linked to option 3(b).  During the final stages of the research we have asked our experts and a selected number of law enforcement and prosecution authorities to provide an estimate of the increase in prosecution activity.  Because of the differences in Member States systems, we have aggregated them to reach the estimated listed below.

If prosecution activity increases by

- 0.1%, the financial impact of the measure would be €2,820,304,143 x 0.1%= €2,820,304

- 0.5%, the financial impact of the measure would be €2,820,304,143 x 0.5%= €14,101,521

- 1%, the financial impact of the measure would be €2,820,304,143 x 1%= €28,203,041

The total cost of Option 3(b) would therefore be a minimum of €90,000 + €2,8 million = €2,9 million per annum.

Right to remain silent, right not to cooperate and privilege against self-incrimination

Option 1 – Status Quo

The quantifiable financial impacts of the baseline scenario (status quo) are linked to (i) the current costs no incurred by evidence not being removed during the proceedings and (ii) those linked to retrial, which are equivalent to any cost incurred by option 3(a).  Furthermore, extra costs could be incurred by procedure being brought against a Member State, having the potential to be receivable by the ECtHR.  As an example, it is estimated that the Cadder case in Scotland cost in excess of €175,000[105].

Option 2 – Non-legislative option

Set up a system to exchange best practice to develop trust between lawyers, judicial and law enforcement authorities to develop understanding of each other’s role and reduce forcing suspects or accused persons to speak or cooperate. Draw up guidelines on the way in which this right should be understood.  Encourage the training of law enforcement officer and judicial authorities staff

The cost of Option 2 would be similar for each issue.  Consequently, the cost of this option can be estimated to reach €7,846,278.  If Option 2 is chosen for each of the research stands however, the total cost would be significantly lower than €7,846,278 x 4 = €31,382,112 as some significant synergies and saving would take place.

Option 3(a) – Legislative option (ECHR standards)

Establish the general principle of the right to silence, the right not to cooperate and the right not to be compelled to produce self-incriminatory evidence (not allowing improper pressure from police or judicial authorities).

Allow for some exceptions:

(i) by allowing adverse inferences to be drawn from exercising such right in exceptional cases according to the ECtHR case law;

- (ii) for the right not to cooperate: by allowing for some very specific exceptions, according to the ECtHR case law (e.g. blood samples, bodily tissue for DNA testing).

+ Remedies: minimum rules on the consequences of breach of these rights.

Costs relating to this option refer to (i) increased costs of the prosecution and (ii) costs of retrials costs

Costs of the removal of evidence at the trial

In the case where evidence is removed, one can assume that this will also have an incidence on the cost of the prosecution.  The scale of this increase is impossible to set out in detail, as this will depend on the complexity of each case.  Assuming that the average case where the evidence is removed leads to an increase of two hours of the prosecution time, the inadmissibility of the evidence in cases where the right to remain silent is breached would be as follows.

Column A – Member State

Column B – Gross annual salary of a Public Prosecutor (CEPEJ)[106]

Column C – Gross weekly salary based on 47 weeks worked per annum (Column B /47)

Column D – Gross hourly salary based on 40 hours worked per week (Column C /40)

Column E – Extra cost of prosecution based on 1 day (Column D x 8 hours)

Column F – Number of criminal cases - CEPEJ

Column G – Costs for 1/10,000 scenario ((Column F / 10,000) x Column E)

Column H – Costs for 1/50,000 scenario ((Column F / 50,000) x Column E)

Column I – Costs for 1/100,000 scenario ((Column F / 100,000) x Column E)

Table D8

A EU jurisdiction || B Gross annual salary of a Public Prosecutor || C Gross salary per week (number of weeks worked = 47 per annum) || D Gross salary per hour (40 hours per week) || E Extra cost of prosecution of not admitting evidence where the right to remain silent has been breached || F Number of criminal cases || G 1 / 10,000 cases || H 1 / 50,000 cases || I 1 / 100,000 cases

Austria ||        50,653 || 1,078 || 27 ||             216 || 60,726 ||  1,309 || 262 || 131

Belgium ||        62,367 || 1,327 || 33 ||             265 || 189,716 ||  5,035 || 1,007 || 503

Bulgaria ||        10,230 || 218 || 5 ||               44 || 118,262 ||  515 || 103 ||   51

Croatia ||        30,396 || 647 || 16 ||             129 || 110,524 ||  1,430 || 286 || 143

Cyprus ||        32,942 || 701 || 18 ||             140 || 117,495 ||  1,647 || 329 || 165

Czech Republic ||        19,632 ||   418 ||  10 ||          84 ||   97,675 ||  816 || 163 ||   82

Estonia ||        15,108 || 321 || 8 ||               64 || 48,359 ||  311 ||   62 ||   31

Finland ||        45,048 || 958 || 24 ||             192 || 59,683 ||  1,144 || 229 || 114

France ||        40,660 || 865 || 22 ||             173 || 1,061,097 ||   18,359 || 3,672 || 1,836

Germany ||        41,127 || 875 || 22 ||             175 || 1,181,995 ||   20,686 || 4,137 || 2,069

Greece ||        32,704 || 696 || 17 ||             139 || 195,929 ||  2,727 || 545 || 273

Hungary ||        16,852 || 359 || 9 ||               72 || 269,691 ||  1,934 || 387 || 193

Ireland ||      52,290 * || 1,113 || 28 ||             223 || 77,625 ||  1,727 || 345 || 173

Italy ||        50,290 || 1,070 || 27 ||             214 || 1,607,646 ||   34,404 || 6,881 || 3,440

Latvia ||        13,524 || 288 || 7 ||               58 || 9,959 || 57 ||   11 ||  6

Lithuania ||        12,529 || 267 || 7 ||               53 || 81,277 ||  433 ||   87 ||   43

Luxembourg ||        78,483 || 1,670 || 42 ||             334 || 14,579 ||  487 ||   97 ||   49

Malta ||      35,699 * || 760 || 19 ||             152 || 19,613 ||  298 ||   60 ||   30

Netherlands ||        54,036 || 1,150 || 29 ||             230 || 441,911 ||   10,161 || 2,032 || 1,016

Poland ||        20,736 || 441 || 11 ||               88 || 1,111,772 ||  9,810 || 1,962 || 981

Portugal ||        35,699 || 760 || 19 ||             152 || 115,466 ||  1,754 || 351 || 175

Romania ||        25,750 || 548 || 14 ||             110 || 171,480 ||  1,879 || 376 || 188

Slovakia ||        26,585 || 566 || 14 ||             113 || 41,189 ||  466 ||   93 ||   47

Slovenia ||        34,858 || 742 || 19 ||             148 || 90,205 ||  1,338 || 268 || 134

Spain ||        47,494 || 1,011 || 25 ||             202 || 1,336,505 ||   27,011 || 5,402 || 2,701

Sweden ||        52,290 || 1,113 || 28 ||             223 || 91,000 ||  2,025 || 405 || 202

UK ||        33,515 || 713 || 18 ||             143 || 1,096,664 ||   15,640 || 3,128 || 1,564

Total (BE, BG, CY, EE, ES, HR, IE, LT, LV, NL, PL, SE, UK) || || || || || ||   75,803 ||   15,161 || 7,580

The total cost of the removal of evidence at the trial stage is thus estimated between €7,500 and €75,800 per annum for Member States where the situation is known. 

Costs of retrials

The further remedy (re-trail) would bear re-trail costs.  The table below provides the costs of re-trial based on the assumption that one in 10,000, 50,000 and 100,000 cases will lead to a retrial.

Table D9

A - EU jurisdiction || Cost per case || 1/10,000 || 1/50,000 || 1/100,000

Number of cases leading to a retrial || cost of retrial || Number of cases leading to a retrial || cost of retrial || Number of cases leading to a retrial || cost of retrial

Austria ||          3,259 ||                  6 ||              19,790 ||                  1 ||            3,958 ||                  1 ||            1,979

Belgium ||          1,359 ||                19 ||              25,787 ||                  4 ||            5,157 ||                  2 ||            2,579

Bulgaria ||             617 ||                12 ||                7,294 ||                  2 ||            1,459 ||                  1 ||               729

Croatia ||          1,243 ||                11 ||              13,735 ||                  2 ||            2,747 ||                  1 ||            1,373

Cyprus ||             186 ||                12 ||                2,181 ||                  2 ||               436 ||                  1 ||               218

Czech Republic ||          2,306 ||                10 ||              22,522 ||                  2 ||            4,504 ||                  1 ||            2,252

Estonia ||             360 ||                  5 ||                1,742 ||                  1 ||               348 ||                  0 ||               174

Finland ||          2,647 ||                  6 ||              15,799 ||                  1 ||            3,160 ||                  1 ||            1,580

France ||          1,444 ||              106 ||            153,170 ||                21 ||          30,634 ||                11 ||          15,317

Germany ||          1,629 ||              118 ||            192,501 ||                24 ||          38,500 ||                12 ||          19,250

Greece ||          1,359 ||                20 ||              26,631 ||                  4 ||            5,326 ||                  2 ||            2,663

Hungary ||             625 ||                27 ||              16,868 ||                  5 ||            3,374 ||                  3 ||            1,687

Ireland ||          1,245 ||                  8 ||                9,667 ||                  2 ||            1,933 ||                  1 ||               967

Italy ||          1,234 ||              161 ||            198,339 ||                32 ||          39,668 ||                16 ||          19,834

Latvia ||          2,410 ||                  1 ||                2,400 ||                  0 ||               480 ||                  0 ||               240

Lithuania ||             404 ||                  8 ||                3,287 ||                  2 ||               657 ||                  1 ||               329

Luxembourg ||             827 ||                  1 ||                1,205 ||                  0 ||               241 ||                  0 ||               121

Malta ||             340 ||                  2 ||                   667 ||                  0 ||               133 ||                  0 ||                 67

Netherlands ||          1,457 ||                44 ||              64,393 ||                  9 ||          12,879 ||                  4 ||            6,439

Poland ||             798 ||              111 ||              88,731 ||                22 ||          17,746 ||                11 ||            8,873

Portugal ||          2,978 ||                12 ||              34,381 ||                  2 ||            6,876 ||                  1 ||            3,438

Romania ||          1,347 ||                17 ||              23,091 ||                  3 ||            4,618 ||                  2 ||            2,309

Slovakia ||          2,186 ||                  4 ||                9,002 ||                  1 ||            1,800 ||                  0 ||               900

Slovenia ||          1,284 ||                  9 ||              11,580 ||                  2 ||            2,316 ||                  1 ||            1,158

Spain ||             813 ||              134 ||            108,677 ||                27 ||          21,735 ||                13 ||          10,868

Sweden ||          3,980 ||                  9 ||              36,222 ||                  2 ||            7,244 ||                  1 ||            3,622

UK ||             837 ||              110 ||              91,752 ||                22 ||          18,350 ||                11 ||            9,175

Total (minus AT,FI,FR,HU) || ||              837 ||            975,788 ||              167 ||        195,158 ||                84 ||          97,579

The total costs of retrials under this option would thus be between € 98,000 and € 980,000 per annum.

The total cost of option 3(a) would therefore be between € 105,500 and €1.05 million per annum.

Option 3(b) – Legislative option (beyond ECHR standards)

Similar to option Option 3 (a) but with exceptions drafted in a narrower way than ECtHR /no exceptions

This option would lead to an increase in the costs at the investigative stage, which can be translated into an increase in the cost of the prosecution.  This is calculated below.

Increased costs of prosecution

The table below provides figures on the total annual budget allocated for public prosecution.  Figures are taken from the CEPEJ report (2012), providing 2010 data.  Figures followed by an asterisk (*) denote Member States for which the data was not provided.  We have calculated the public annual budget allocated to public prosecution per inhabitant where possible and extrapolated the data based on the average (€11.9).  Overall, we estimate that the total annual budget allocated to public prosecution in the EU (with the exception of Denmark and the addition of Croatia) is €6,737,848,858 (€ 6.8 billion)

Table D10

EU jurisdiction || Public annual budget allocated to public prosecution || Population || Public annual budget allocated to public prosecution per inhabitant

Austria ||              100,231,686* ||          8,404,252 || n/a

Belgium ||              130,608,155* ||        10,951,266 || n/a

Bulgaria ||                79,203,203 ||          7,504,868 || 10.6

Croatia ||                41,296,176 ||          4,412,137 || 9.4

Cyprus ||                15,964,412 ||             804,435 || 19.8

Czech Republic ||                83,446,289 ||        10,532,770 || 7.9

Estonia ||                  9,135,614 ||          1,340,194 || 6.8

Finland ||                42,937,000 ||          5,375,276 || 8.0

France ||              775,787,302* ||        65,048,412 || n/a

Germany ||              974,994,666* ||        81,751,602 || n/a

Greece ||              134,885,155* ||        11,309,885 || n/a

Hungary ||              102,321,320 ||          9,985,722 || 10.2

Ireland ||                43,854,000 ||          4,480,858 || 9.8

Italy ||           1,249,053,619 ||        60,626,442 || 20.6

Latvia ||                15,913,545 ||          2,229,641 || 7.1

Lithuania ||                29,555,000 ||          3,244,601 || 9.1

Luxembourg ||                  6,104,361* ||             511,840 || n/a

Malta ||                  2,569,000 ||             417,617 || 6.2

Netherlands ||              615,642,000 ||        16,655,799 || 37.0

Poland ||              312,514,570 ||        38,200,037 || 8.2

Portugal ||              113,901,622 ||        10,636,979 || 10.7

Romania ||              162,428,333 ||        21,390,000 || 7.6

Slovakia ||                63,702,886 ||          5,435,273 || 11.7

Slovenia ||                19,263,376 ||          2,050,189 || 9.4

Spain ||              550,433,943* ||        46,152,926 || n/a

Sweden ||              127,316,425 ||          9,415,570 || 13.5

UK ||              934,785,200 ||        62,641,000 || 14.9

Total ||           6,737,848,858 || ||

Total MS where adverse inferences can be drawn (BE, CY, FI, FR, IE, LV, NL, SE, UK) ||           2,702,808,039 || ||

While it is not possible to assess the exact financial impact of the increased prosecution costs, we can provide some likely scenarios. Below we provide different scenarios.  It is very difficult to assess the increase of prosecution activity linked to option 3(b).  During the final stages of the research we have asked our experts and a selected number of law enforcement and prosecution authorities to provide an estimate of the increase in prosecution activity.  Because of the differences in Member States systems, we have aggregated them to reach the estimated listed below.

In the event where the measure increases prosecution activity in those MS where the burden of proof can currently be reversed by

- 0.5%, the financial impact of the measure would be €2,702,808,039 x 0.5%= €13,514,040

- 1%, the financial impact of the measure would be €2,702,808,039 x 1%= €27,028,080

- 2%, the financial impact of the measure would be €2,702,808,039 x 2%= €54,056,161

While it is difficult to assess the exact increase in the costs at the investigative stage, one can assume that these will be greater under Option 3(b). Assuming that the costs of the prosecution increase by 2%, the cost of this option would be €54 million.

Taking into account the “worst scenario”, the total cost of Option 3(b) would be €54 million per annum.

Right to be tried in one’s presence

Option 1 – Status Quo

The financial impacts of the status quo can be assessed as (i) costs not incurred linked to some Member States having too low a threshold for trial to be held in the absence of the defendant and (ii) extra costs could be incurred by procedure being brought against a Member State, having the potential to be receivable by the ECtHR.  As an example, it is estimated that the Cadder case in Scotland cost in excess of €175,000[107].

Option 2 – Non legislative option

Set up a system to exchange best practice to develop trust between lawyers, judicial and law enforcement authorities to develop understanding of each other’s role and reduce forcing suspects or accused persons to speak or cooperate. Draw up guidelines on the way in which this right should be understood.  Encourage the training of law enforcement officer and judicial authorities staff

The cost of Option 2 would be similar for each issue.  Consequently, the cost of this option can be estimated to reach €7,846,278.  If Option 2 is chosen for each of the research stands however, the total cost would be significantly lower than €7,846,278 x 4 = €31,382,112 as some significant synergies and saving would take place.

Option 3(a) – Legislative option (ECHR standards)

Establish the general principles according to the ECtHR case law (person charged with a criminal offence is entitled to take part in the hearing, and can be tried in absentia only if he had an opportunity to waive his right to be present).

+ Remedy: right to re-open the criminal proceedings (not only the right to appeal) if the in absentia rules were not respected.

Costs linked to retrials

This Option would lead to an increase in the number of re-trials.  Currently, in four jurisdictions (BE, BG, HU and LV) a defendant who was not tried in his present may not obtain a fresh determination of the merits of the charge from the court which heard the case. Assuming that 1 in 1,000 trials is held in absentia, and using the figures calculated above, the table below shows that the total cost of this measure would amount to €523,480 per annum.

Table D11

EU jurisdiction || Cost per case || Number of cases leading to a retrial || Cost of retrial

Austria ||          3,259 ||   60 ||  197,900

Belgium ||          1,359 || 190 ||  257,870

Bulgaria ||             617 || 120 || 72,940

Croatia ||          1,243 || 110 ||  137,350

Cyprus ||             186 || 120 || 21,810

Czech Republic ||          2,306 || 100 ||  225,220

Estonia ||             360 ||   50 || 17,420

Finland ||          2,647 ||   60 ||  157,990

France ||          1,444 ||  1060 ||   1531,700

Germany ||          1,629 ||  1180 ||   1925,010

Greece ||          1,359 || 200 ||  266,310

Hungary ||             625 || 270 ||  168,680

Ireland ||          1,245 ||   80 || 966,700

Italy ||          1,234 ||  1610 ||   1,983,390

Latvia ||          2,410 ||   100 || 24,000

Lithuania ||             404 ||   80 || 32,870

Luxembourg ||             827 ||   10 || 12,050

Malta ||             340 ||   20 || 6,670

Netherlands ||          1,457 || 440 ||  643,930

Poland ||             798 ||  1110 ||  887,310

Portugal ||          2,978 || 120 ||  343,810

Romania ||          1,347 || 170 ||  230,910

Slovakia ||          2,186 ||   40 || 90,020

Slovenia ||          1,284 ||   90 ||  115,800

Spain ||             813 ||  1340 ||   1,086,770

Sweden ||          3,980 ||   90 ||  362,220

UK ||             837 ||  1100 ||  917,520

Total for BE, BG, HU and LV || || || 523,480

Option 3(b) - Legislative option (beyond ECHR standards)

Option 3 (a) but increasing the safeguards to ensure that even less trials take place without the presence of the accused.

Costs linked to ensuring the presence of the defendant at his or her trial

There are no data to calculate the costs of Option 3(b), however, by limiting the cases in which in absentia proceedings can take place, one can assume that the costs of this option would be linked to the resources used to ensure that a suspect or accused is physically brought to trial.  Assuming that each intervention would occupy 3 police officers for 3 hours, the total cost of this measure if we accept the scenario where between 0.5% and 2% of cases would be affected would be between €5.5 million and €22 million per annum.

It is not possible to know how many cases will be affected so we only provide a ranged figure as result of the costs for this option

Column A – Member State

Column B – GDP per capita in PPS (base point – Eurostat)

Column C – Average salary of a police officer (sources – Belgium – SPF interieur figures, Home Office figures)

Column D – Average salary of a police officer per GDP base point – where known (Column C / Column B)

Column E – Salary of a police officer per annum (Column B x average column C)

Column F – Cost of police officer for 3 hours (Column E / 47 working weeks / 40 hours *3 hours per intervention)

Column G – Number of criminal cases

Column H – I - J – Cost of the different scenarii

Table D12

A EU jurisdiction || B GDP/capita in PPS (2010) - EUROSTAT (base point) || C Average salary of a police officer  (where known) || D Average salary of a police officer per base point || E Salary of a police officer per annum (average per base pint x base point) || F Cost of police officer for 3 hours (Column E / 47 working weeks / 40 hours *3 hours per intervention) || G Number of criminal cases || H 0.5% cases || I 1% cases || J 2% cases

Austria || 126 || || || 28,989 || 139 || 60,726 ||   42,138 || 84,275 || 168,550

Belgium || 119 || 16,872 || 142 || || 131 || 189,716 ||   76,617 ||  153,234 || 306,469

Bulgaria || 44 || || || 10,123 ||   48 || 118,262 ||   28,656 || 57,313 || 114,626

Croatia || 61 || || || 14,035 ||   67 || 110,524 ||   37,129 || 74,258 || 148,515

Cyprus || 99 || || || 22,777 || 109 || 117,495 ||   64,059 ||  128,117 || 256,235

Czech Republic || 80 || || || 18,406 ||   88 || 97,675 ||   43,033 || 86,065 || 172,130

Estonia || 64 || || || 14,725 ||   70 || 48,359 ||   17,044 || 34,089 ||   68,178

Finland || 115 || || || 26,459 || 127 || 59,683 ||   37,798 || 75,597 || 151,193

France || 108 || || || 24,848 || 119 || 1,061,097 || 631,107 ||  1,262,213 || 2,524,426

Germany || 118 || || || 27,149 || 130 || 1,181,995 || 768,107 ||  1,536,213 || 3,072,426

Greece || 90 || || || 20,707 ||   99 || 195,929 ||   97,110 ||  194,221 || 388,441

Hungary || 65 || || || 14,955 ||   72 || 269,691 ||   96,539 ||  193,078 || 386,157

Ireland || 128 || || || 29,450 || 141 || 77,625 ||   54,719 ||  109,437 || 218,875

Italy || 101 || || || 23,238 || 111 || 1,607,646 || 894,202 ||  1,788,404 || 3,576,808

Latvia || 51 || || || 11,734 ||   56 || 9,959 ||  2,797 ||   5,594 ||   11,188

Lithuania || 57 || || || 13,114 ||   63 || 81,277 ||   25,513 || 51,027 || 102,053

Luxembourg || 271 || || || 62,350 || 298 || 14,579 ||   21,758 || 43,516 ||   87,032

Malta || 83 || || || 19,096 ||   91 || 19,613 ||  8,965 || 17,930 ||   35,860

Netherlands || 133 || || || 30,600 || 146 || 441,911 || 323,676 ||  647,352 || 1,294,703

Poland || 63 || || || 14,495 ||   69 || 1,111,772 || 385,727 ||  771,454 || 1,542,908

Portugal || 80 || || || 18,406 ||   88 || 115,466 ||   50,871 ||  101,741 || 203,483

Romania || 46 || || || 10,583 ||   51 || 171,480 ||   43,441 || 86,881 || 173,762

Slovakia || 74 || || || 17,026 ||   82 || 41,189 ||   16,786 || 33,571 ||   67,142

Slovenia || 85 || || || 19,556 ||   94 || 90,205 ||   42,225 || 84,451 || 168,901

Spain || 100 || || || 23,007 || 110 || 1,336,505 || 736,028 ||  1,472,056 || 2,944,113

Sweden || 123 || || || 28,299 || 135 || 91,000 ||   61,641 ||  123,282 || 246,564

UK || 112 || 35,657 || 318 || || 171 ||  1,096,664 || 936,000 ||  1,872,001 || 3,744,001

Total || || || 230 || || || || 5,543,685 || 11,087,371 || 22,174,741

ANNEX VII Examples of when lack of mutual trust can hinder judicial cooperation

Ø The subject matter of a number of the recent preliminary references from national courts to the European Court Justice (ECJ) on the Framework Decision on the European arrest warrant[108] (FD-EAW) are illustrative of the continuing gaps in mutual trust between Member States that can be closed by EU common minimum standards of procedural rights as follows:

Ø In the recent Radu case[109] (judgment on the 29 January 2013) the Romanian court of appeal demonstrated by the nature and breadth of its questions to the ECJ that they did not have the required levels of trust in the EAW system. The wide-ranging questions were about the compatibility of the arrest of a person and the execution of an EAW with fair trial rights and rights to liberty in the EU Charter of Fundamental Rights (the Charter) and in the European Convention on Human Rights (the ECHR) and about the adequacy and compatibility of transposition of the FD-EAW in both issuing and executing Member States. The Court ultimately interpreted the questions in a narrow manner and ruled that judicial authorities cannot refuse to execute an EAW on the ground that the requested person was not heard in the issuing Member State before that arrest warrant was issued. However the case illustrates that in very recent times judicial authorities are questioning the levels of procedural rights for suspects and accused persons in other Member States with the potential to lead to considerable delay in the ultimate decision on judicial cooperation.

Ø The questions raised by the Spanish constitutional court in the Melloni case[110] (judgment 26th February 2013) on the obligations of an executing judicial authority where there are different standards of protection in respect of In Absentia trials (despite this issue having been harmonised in respect of EAW cases in the Framework Decision on In Absentia judgments[111]) shows a lack of trust in the standards of protection of the presumption of innocence  that has the potential to delay judicial cooperation and did in this case. The ECJ concluded that the difference in the standard of protection between the issuing and the executing Member State was not a reason to refuse the surrender as long as certain minimum standards were respected. As a consequence, the person was surrendered, but only after a serious delay following several court proceedings.

Ø The issue of the application of the rule of speciality (which prohibits prosecution for prior offences other than those in the warrant)  has been the subject of a very recent ECJ case Jeremy F[112] (judgment on 4 April 2013) and was also the subject of the case of Leymann and Pustovarov[113] (judgment on 1 December 2008).  In the recent case of Melvin West[114] (judgment 28 June 2012) the issue raised was about consent to onward surrender to another Member State. In the case of Advocaten voor de Wereld[115] (judgment of 3 May 2007) the ECJ was asked to consider the compatibility of the non-verification of dual criminality for listed offences with the principle of equality and non-discrimination. The fact that all of these issues continue to be raised at ECJ level indicates the distance we still have to travel in terms of achieving mutual trust. This is because the FD-EAW gave Member States a number of options, which were largely not availed of, to have a very high level of judicial cooperation. These included the possibility to dispense with the requirement of dual criminality altogether (Article 2.4) the possibility to waive the rule of speciality entirely in dealings with other Member State (Article 27.1) and the possibility not to require consent for onward surrender to another Member State (Article 28.1).  Ensuring minimum standards of procedural rights will help to create the conditions for Member States to be happy to enhance cooperation by availing of these possibilities. The ECJ would no longer be required to deal with issues such as dual criminality, speciality and onward surrender as they would not arise where mutual trust is optimum.

Ø It is clear from the experience with the EAW that lack of mutual trust can result in complex and long-drawn out investigations into the systems of other MS because of procedural rights issues raised at first instance and on appeal. This creates delays that can ultimately prejudice the resolution of cases for all parties involved, despite the fact that in the vast majority of EAW cases the ultimate decision (unless an agreed refusal ground applies) is to surrender the person.  A high-profile recent illustration of this scenario is the Swedish-UK case of Mr. Julian Assange[116] whose surrender was ultimately confirmed by the UK Supreme Court in June 2012, a year and a half after his initial arrest in December 2010 in the UK pursuant to the EAW issued by Sweden. Mr. Assange raised wide-ranging issues including the legitimacy of the authority that issued the EAW, dual criminality, whether a decision to prosecute had been taken by the Swedish authorities and the proportionality of the request.  The lack of common minimum standards of procedural rights can be exploited to lead to challenges that have the potential to considerably delay judicial co-operation.

Ø It is the case that in their implementation of the EAW FD, a number of Member States have chosen to go beyond the EAW-FD in providing for more stringent rules for surrender of their own citizens, indicating a level of mistrust that the procedural rights measures will help to address. One example is the case of Klaas Karel Faber[117], a former Member of the Waffen SS in the Netherlands, who was sought by the Netherlands from Germany pursuant to an EAW following his convicted to life imprisonment for murders. In 2011, a German court refused to surrender Faber to serve his sentence in the Netherlands on the grounds that his consent to surrender was required, thus adding an element that is not in the FD-EAW.

Ø The case of Gary Mann[118] shows the effect that a lack of EU minimum procedural safeguards can have on intra-EU judicial cooperation. The case relates to the execution of a Portuguese European arrest warrant by UK courts, for the surrender of Gary Mann, a British citizen, which took more than 14 months (the Framework Decision on the EAW provides for a sixty-day deadline) and involved five decisions by UK courts. The main issue raised was inadequate legal advice, since Mann and eleven other defendants were represented by only one lawyer. In addition, Mann was unable properly to instruct his lawyer due to the lack of time before the hearing. Following his arrest, trial and conviction that took place in less than 48 hours, he was finally sentenced to two years' imprisonment for his role in a riot at the Euro 2004 tournament. The case clearly shows that the execution of the EAW will happen much more swiftly if the executing judicial authority can be confident that there are minimum standards of procedural safeguards that are enforceable across the EU.

Ø The case of Deborah Dark shows that insufficient trust in the standards of protection of fair trial rights (lack of notification of the appeal, no legal representation during the appeal hearing and lack of information of the conviction, delay) may hinder effective judicial co-operation. In 1989 Deborah Dark was arrested in France on suspicion of drug related offenses but the court acquitted her of all charges. In 1990, she was convicted and sentenced to prison on appeal without herself or her French lawyer being notified.  In 2005, an EAW was issued by the French authorities. In 2008 and 2009, Ms. Dark was arrested successively in Spain and in UK, and at the extradition hearing both of the national courts refused to extradite Ms. Dark to France. In May 2010, France finally agreed to remove the warrant. This case shows that there work remains to be done on minimum procedural rights to ensure the effective right of a suspect to fair trial and the essential confidence of  judicial authorities in the systems of other Member States.

Ø The impact of concerns including those relating to the presumption of innocence in undermining mutual trust are illustrated by a recent English Appeal court case of Sofia City Court v Dimintrinka Atanasova-Kalaidzheiva of 2011[119]. The UK courts refused to execute an EAW at first instance and on appeal on the basis that they had doubts that a fair trial was possible in this particular case and were not satisfied about the independence of the investigation and prosecution process in Bulgaria.  Judicial authorities must be confident that the key right to be presumed innocent that underpins a fair trial is guaranteed

ANNEX VIII Glossary of main legal terms used

Appeal (in connection to retrial) – both mechanisms seek re-evaluation of a judicial decision by a different court than the one which took that decision. The main differences between the two are:

· as a general rule, an appeal is limited to a re-examination of the case by a higher court in terms of questions of law and, contrary to a retrial, no new examination of evidence takes place. A retrial means, as defined by the ECtHR, a procedure whereby a fresh determination of the merits of the charges is ensured, and this is not possible under the general rules of an appeal.

· an appeal intervenes only ex post, i.e., only after the first judgment has been delivered, which in relation to presumption of innocence means that the accused will only have the right to a fair trial in the higher court, and thus the right to have his case submitted to a double degree of jurisdiction does not exist in practice.

'In absentia' decisions – judicial decisions taken without the presence of the suspect or accused person concerned.

Natural persons (as opposed to legal persons) – a natural person is any human being, with legal capacity commencing from the time of birth. A legal person is an association of people or special-purpose fund (e.g. a foundation) whose legal personality is recognized by law.

Rebuttable (in relation to a presumption) - possibility to contradict a presumption (in particular, a presumption of guilt), by means of, e.g., providing enough evidence to prove innocence.

Strict liability offences – offences for which the mere proof of certain objective facts is sufficient to prove guilt (as opposed to the general rule according to which criminal liability also requires an element of intention).

Suspect - someone who is suspected of having committed a criminal offence but has not yet been formally charged; accused person - someone who has been formally charged with an offence. Their rights are different according to their status in accordance with national law. There is no EU definition of these notions.

[1] A glossary of the main legal terms used in this Impact Assessment is available in Annex VIII. A suspect is someone who is suspected of having committed a criminal offence but has not yet been formally charged. An accused person is someone who has been formally charged with an offence. Their rights are different according to their status in accordance with national law. There is no EU definition of these notions. However, both categories are entitled to be presumed innocent.

[2] This impact assessment therefore does not cover administrative proceedings.

[3] CSES study referred to under point 3.2.

[4] Article 6(2) ECHR: "Everyone charged with a criminal offence shall be presumed innocent until proved guilty by law."

[5] Article 11(1) of the Universal Declaration of Human Rights safeguards the principle as follows: "Everyone charged with a penal offence has the right to be presumed innocent until proved guilty according to law in a public trial at which he has had all the guarantees necessary for his defence." Similar principles are laid down in Article 14 of the United Nations International Covenant on Civil and Political Rights.

[6] See Annex V. Similarly to some Common law EU Member States, the US legal system does not enshrine the presumption of innocence in the US Constitution; however it is widely recognised to be implied in the 5th, 6th and 14th amendments. Moreover, the US Supreme Court has developed the so-called Miranda doctrine. Under this doctrine, prior to any questioning during custodial investigation, the person must be warned that he has a right to remain silent, that any statement he gives may be used as evidence against him, and that he has the right to the presence of an attorney, either retained or appointed.

[7] Barberà, Messegué and Jabardo v. Spain, Application No.s 10588/83, 10589/83 and 10590/83, judgment of 6.12.1988.

[8] E.g., Murray v. UK (application 18731/91, judgement of 8.2.1996), Saunders v. UK (Application 19187/91, judgement of 17.12.1996).

[9] E.g., Kudla v. Poland (application 30210/96, judgment of 26 October 2010).

[10] COM(2011) 326 final, 8.6.2011. The themes of the Green Paper were pre-trial detention and mutual recognition of custodial and non-custodial decisions. The Commission received 81 replies from Member States, civil society and NGOs. A summary of the replies has been published on the website of the Commission. http://ec.europa.eu/justice/newsroom/criminal/opinion/110614_en.htm. See also Framework Decision 2009/829/JHA of 23 October 2009 on the application of the principle of mutual recognition to decisions on supervision measures as an alternative to provisional detention (OJ L 294, 11.11.2009, p. 20).

[11] E.g., Colozza v. Italy (application 9024/80, judgment of 12.2.1985).

[12] Article 6(1) reads: “In the determination of his civil rights and obligations or of any criminal charge against him, everyone is entitled to a fair and public hearing within a reasonable time by an independent and impartial tribunal established by law. Judgment shall be pronounced publicly but the press and public may be excluded from all or part of the trial in the interests of morals, public order or national security in a democratic society, where the interests of juveniles or the protection of the private life of the parties so require, or to the extent strictly necessary in the opinion of the court in special circumstances where publicity would prejudice the interests of justice.”

[13] See, e.g., Heaney and McGuiness v. Ireland (Application 34720/97, Judgment of 21 December 2000), paragraph 59.

[14]             See, inter alia, Case C-301/04 P Commission v SGL Carbon [2006] ECR I-5915; Case T-112/98 Mannesmannröhren-Werke v Commission [2001] ECR II-732, Case C-374/87 Orkem v Commission [1989] ECR 3343. The recognition of more limited rights of legal persons also finds support in international practice (e.g. in the US the privilege against self-incrimination is recognised solely for natural persons, with legal persons having no such right: United States v. Kordel, 397, U.S. 1 (1970) and Braswell v United States, 487 U.S. 99 (1988)).

[15] "Study on Analysis of the future of mutual recognition in criminal matters in the European Union", G.Vernimmen – Van Tiggelen and Laura Surano, Call for tenders JLS D3/2007/03, European Commission, 20 November 2008, para. 18.

[16] OJ C 115, 4.5.2010, p. 1.

[17] OJ C 291, 4.12.2009, p. 1.

[18] The first two measures have already been adopted: Directive 2010/64/EU of the European Parliament and of the Council on the right to interpretation and translation in criminal proceedings (OJ L 280, 26.10.2010, p. 1); Directive 2012/13/EU of the European Parliament and of the Council on the right to information in criminal proceedings (OJ L 142, 1.6.2012, p. 1. The third measure, access to a lawyer and legal aid, was split into two parts and a Directive on the right to access to a lawyer has also been adopted: Directive 2013/48/EU of the European Parliament and of the Council of 22 October 2013 on the right of access to a lawyer in criminal proceedings and in European arrest warrant proceedings, and on the right to have a third party informed upon deprivation of liberty and to communicate with third persons and with consular authorities while deprived of liberty (OJ L 294, 6.11.2013, p. 1) (COM(2011) 326 final, 8.6.2011); this Directive also includes the fourth measure, on the right for a detained person to communicate with family members, employers and consular authorities; The fifth measure on the protection of vulnerable persons suspected or accused in criminal proceedings and the measure on legal aid (the part of the third measure not included in Directive 2013/48/EU) are presented as a package together with the present initiative. Green Paper Strengthening mutual trust in the European judicial area – A Green Paper on the application of EU criminal justice legislation in the field of detention, COM/2011/0327 final (http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:2011:0327:FIN:EN:PDF).

[19] Directive 2012/29/EU of the European Parliament and of the Council establishing minimum standards on the rights, support and protection of victims of crime, and replacing Council Framework Decision 2001/220/JHA (OJ L 315, 14. 11. 2012, p. 57)

[20] COM(2006) 174 final, 26.4.2006.

[21] Annex II – list of stakeholders consulted.

[22] Available at: http://www.fairtrials.net/publications/policy-and-campaigns/defence-rights-in-the-eu-report/.

[23] Available at http://www.justice.org.uk/resources.php/328/european-arrest-warrants

[24] Ed Cape, Zaza Namoradze, Roger Smith and Taru Spronken, Intersentia, 2010.

[25] Ed Cape and Zaza Namoradze, LARN, 2012.

[26] See footnote 12.

[27] Available at the ECtHR's website; http://www.echr.coe.int/echr/homepage_EN

[28] See Annex IV.

[29] As regards what are the key requirements of presumption of innocence and what are other rights closely related to it, see introduction to this Impact Assessment.

[30] See CSES study, p. 10.

[31] Member States' potential savings owing to a reduction in a number of appeals, condemnations by the ECtHR, or delays in judicial cooperation proceedings cannot be estimated with any statistical precision due to lack of Member States' data on costs per case. Only indicative qualitative expectations in non-numerical terms can therefore be provided based on stakeholders' judgments. However, as an example of the cost of a case being brought through the domestic systems and ultimately before the ECtHR, it is estimated that the Cadder case in Scotland on insufficient legal representation cost in excess of €175,000 (see further explanation in Annex VI and its Table D2).

[32] See e.g. Christou et al, European Cross Border Justice: A Case Study of the EAW, The AIRE Centre, 2010.

[33] Statement of ECBA on the Green Paper on Presumption of Innocence of 26 April 2006, July 2006, pp 5-6.

[34] See Colozza v. Italy (application 9024/80, judgment of 12.2.1985).

[35] See "Effective Criminal Defence in Europe", p. 13.

[36] Article 1(3) of the EAW Framework Decision presupposes that the underlying procedure should respect the principle in articles 47 and 48 of the Charter, but it is not a ground for refusal of execution of an EAW.

[37] Court of Justice of the European Union, case C-399/11.

[38] [2011] EWHC 2335 (Admin).

[39] Moreover, the absence of common minimum standards on procedural rights may also affect the effectiveness of the criminal proceedings in the issuing Member State. The issuing Member State has no guarantees that the evidence is collected, a house is searched, a person is heard, or assets in a bank account frozen according to a procedure that is in respect of fair trial rights in the executing Member State. This can affect the possibility to rely on such evidence in court in the issuing Member State.

[40] Lord Justice Thomas, the most senior English judge responsible for EAW cases, spoke of judges in the Netherlands who “hold the view that you have expressed here, which is the fact that we have the common area for justice that was put in place with mutual confidence but we know that there are countries where what is on paper is not the actuality. My concern is that you [the Inquiry] might be perceived as looking at this through Anglo-Saxon eyes. [In fact] our views on the problems of EAWs, which arise largely because procedural standards are not common across Europe, are shared by quite a lot of judges.”

[41] Page 330 of the Report.

[42] In their submission to the online survey organized in the framework of the study referred to in point 3.2, one NGO ('JUSTICE') has replied to be in principle in favour of EU action, stating that “although there are few cases (of cross-border criminal proceedings) to show mistrust across borders on (the ground of failures to protect suspects or accuseds’ right to the presumption of innocence), there are clear areas where EU legislation could improve trial standards.” Furthermore, it was thought unlikely that a court in the executing state would refuse to grant an EAW on the basis of any breaches of the right to be presumed innocent and its related fair trial rights, given all the other factors for the court to consider in relation to the trial. It should however be noted that participants in the UK discussion group, organized in the framework of the CSES study referred to in point 3.2, thought the UK courts reject EAWs approximately once a month on Article 6 of the ECHR/fair trial grounds.

[43] See e.g. recent research by JUSTICE, ‘European Arrest Warrants – ensuring an effective defence’, 2012, referred to in point 3.2.

[44] See Minelli v. Switzerland (Application 8660/79, judgment of 25.3.1983).

[45] See Krause v. Switzerland (Application 7986/77, judgment of 13.12.1978.

[46] See Allenet de Ribemont v. France (Application 15175/89, judgment of 10.2.1995).

[47] See table on legal situation in the Member States, in Annex V.

[48] Application 42095/98, judgement of 10 October 2000.

[49] Application 48297/99, judgment of 26 March 2002.

[50] Application 13583/02, judgment of 21 September 2006.

[51] Application 25720/05, judgment of 13 July 2010.

[52] Application 124/04, judgment of 7 February 2012.

[53] Application 44418/07, judgment of 14 June 2011.

[54] Application 18851/07, judgment of 12 April 2012.

[55] Application 13583/02, judgment of 21 September 2006.

[56] Application 36921/07, judgment of 14 June 2011.

[57] Report of Fair Trials International on "Defence Rigths in the EU", p. 50.

[58] Ibid.

[59] Barberà, Messegué and Jabardo v. Spain, paragraph 77.

[60] Telfner v. Austria, Application 33501/96, paragraph 19.

[61] Application 10519/83, judgment of 7.10.1988. This case concerned a person who had passed through customs with cannabis in a suitcase which he declared to be his property; he was convicted under the relevant provision of the Customs Code, which deems anyone carrying in contraband goods (consciously or not) guilty of an offence. Here the Strasbourg Court accepted the respondent state's argument that the strict liability offence was not disproportionate.

[62] See e.g. the very precise rules on reversal of proof set out by the UK higher courts to comply with the standard of Article 6(2) ECHR. They have developed “reasonable limits” criteria for circumstances in which placing a legal burden of proof on the defendant is in the court’s view proportionate. See CSES study, p. 19.

[63] See http://www.theguardian.com/world/2013/jul/31/sture-bergwall-thomas-quick-meeting

[64] Report of Fair Trials International on "Defence Rigths in the EU", p. 61.

[65] See Effective Criminal Defence in Europe, p. 347.

[66] See Effective Criminal Defence in Eastern Europe, p. 137.

[67] Heaney and McGuiness v. Ireland, Application 34720/97, judgment of 21 December 2000.

[68] Ibid.

[69] See Saunders v. UK (Application 19187/91, judgment of 17 December 1996), Funke v. France (Application 10821/84, Judgment of 25 February 1993).

[70] See Condron v. UK, Application 35718/97 (Judgement of 2 May, 2000), paragraph 69.

[71] E.g. ECBA (European Criminal Bar Association).

[72] See e.g. the situation in Germany, where case law has established that "full silence" (as opposed to partial silence) during police questioning may never be held against the accused later in the criminal proceeding. However, it appears from discussions with German stakeholders that in practice the rigour of this position is mitigated by the principle of the free evaluation of evidence that is followed by German judges.

[73] Notwithstanding that there is a system of free admissibility of evidence in Finland, in May 2012 the Finnish Supreme Court ruled that statements obtained where the defendant’s right to a lawyer had been breached could not be used at trial. The Supreme Court held that if the defendant’s right not to incriminate oneself has been breached, this right cannot be corrected simply by allowing the evidence to be presented to the court and the court later on deciding not to give credibility to the statement. The Supreme Court noted that the only way to avoid a violation of the rights of the defendant in these cases is for the incriminating evidence to not be brought to trial at all.

[74] ECtHR Application 54810/00, judgment of 11 July 2006.

[75] On-going study "Inside Police Custody", referred to in point 3.2.

[76] See Colozza v. Italy, paragraph 29.

[77] See Colozza v. Italy, paragraph 29.

[78] See Annex V.

[79] Council Framework Decision 2009/299/JHA enhancing the procedural rights of persons and fostering the application of the principle of mutual recognition to decisions rendered in the absence of the person concerned at the trial (OJ L 81, 27.3.2009, p. 24) amending Framework Decisions 2002/584/JHA on the European arrest warrant and the surrender procedures between Member States, 2005/214/JHA on the application of the principle of mutual recognition to financial penalties, 2006/783/JHA on the application of the principle of mutual recognition to confiscation orders, 2008/909/JHA on the application of the principle of mutual recognition to judgments in criminal matters imposing custodial sentences or measures involving deprivation of liberty for the purpose of their enforcement in the European Union, 2008/947/JHA on the application of the principle of mutual recognition to judgments and probation decisions with a view to the supervision of probation measures and alternative sanctions.

[80] CEPEJ Report, 2012, referred to on p. 37 of the CSES final report.

[81] However, were they to do so this would still not reveal the number of abuses of rights that occurs. The on-line survey carried out by the CSES in the framework of the Study on presumption of innocence show the inevitable gaps between theoretical safeguards and reality, even in those states whose criminal procedures appear on paper to be compliant with ECHR jurisprudence. To this extent the precise scope of the human rights problems concerning presumption of innocence is unknown.

[82] See Annex IV.

[83] In the Netherlands a target has been set out that at least 50% of criminal proceedings will be dealt with by a form of summary procedure for minor offences.

[84] 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.

[85] Eurobarometer 337/2010.

[86] http://ec.europa.eu/justice/news/intro/doc/com_2010_573_4_en.pdf

[87] See introduction, footnote 1.

[88] For details of calculation of all costs for each policy option, see Annex VI.

[89] As pointed out before, Member States' potential savings owing to a reduction in a number of appeals, condemnations by the ECtHR, or delays in judicial cooperation proceedings cannot be estimated with any statistical precision due to lack of Member state data on costs per case. Only indicative qualitative expectations in non-numerical terms can therefore be provided based on stakeholders' judgments. However, as an example of the cost of a case being brought through the domestic judicial system and ultimately before the ECtHR, it is estimated that the Cadder case in Scotland on insufficient legal representation cost in excess of €175,000 (see further explanation in Annex VI and its Table D2).

[90]             Table D3 in Annex VI explains the calculation of monitoring and evaluation costs for of one of the four aspects of presumption of innocence (costs of two full time equivalents). If monitoring and evaluation is done four all four aspects, the costs would be significantly lower than multiplying the costs for one aspect per four (costs of 8 full time equivalents), since significant synergies and savings would take place. The present table includes the costs of four full time equivalents.

[91]             Same reasoning as for monitoring and evaluation. Table D4 in Annex VI explains that costs for training are based on costs per number of training session required. For the present table, it is estimated that one would need the double of training sessions to teach all four aspects of presumption of innocence rather than the quadruple.

[92]             These costs occur only for those Member States which do not yet have a specific remedy in their national law (see Annex VI, table D8).

[93]             These costs occur only for those Member States in which the law permits inferences to be drawn from silence. Calculation based on the assumption that public annual budget allocated to public prosecution will increase by 1% (seen further explanations in Annex VI, table D10).

[94]             These costs occur only for those Member States where a defendant who was not tried in his presence is not entitled to obtain a fresh determination of his case (see Annex IV, table D11).

[95]             OJ C 291, 4.12.2009, p. 1.

[96] 'Derived from the PoI' refers to situations where there are no specific law on the issue but the right and / or the remedy can be derived from the general rules governing the presumption of innocence ('PoI').

[97] The defence is required to lodge a defence statement in solemn proceedings.

[98] In a report by the Law Commission of England and Wales and the Scottish Law Commission in October 2000 on Damages under the Human Rights Act 1998, the E&W Law Commission concluded that although the Strasbourg court’s terminology and concepts with regard to the awarding of damages were different “it frequently reaches very similar results to those reached under the rules of tort”. lawcommission.justice.gov.uk/docs/lc266_damages_under_the_human_rights_act_1998.pdf_

[99] Application 73797/01.

[100] Application 40523/08.

[101] The appeal process involved the refusal of Scottish courts to give permission to appeal and refusal to permit appeal to the Supreme Court. Cadder's representatives eventually succeeded in obtaining special leave to appeal to the Supreme Court under the Scotland Act 1998. Peter Cadder's legal costs throughout this process were borne by legal aid and have been estimated as at least £30,000 (35,000 euros).[101] The prosecution's costs falling directly on the state were likely to have been similar, if not greater, and there would have been further indirect costs, such as court and judges' time and delays caused to other proceedings.  As a consequence the total costs of the Supreme Court decision are estimated at around £150,000 (176,000 euros).(JUSTICE and Maria Fletcher).

[102] Inflation weighed cost of a lawyer with figures taken from the 2007 Study on the Transparency of Costs of Civil Judicial Proceedings in the European Union.

[103] There are no available data on the number of non-criminal cases in the EU.  We have therefore assumed that the share of criminal cases out of total cases is equivalent to the share of criminal legal aid out of all legal aid expenditure = 65%.

[104] The appeal process involved the refusal of Scottish courts to give permission to appeal and refusal to permit appeal to the Supreme Court. Cadder's representatives eventually succeeded in obtaining special leave to appeal to the Supreme Court under the Scotland Act 1998. Peter Cadder's legal costs throughout this process were borne by legal aid and have been estimated as at least £30,000 (35,000 euros).[104] The prosecution's costs falling directly on the state were likely to have been similar, if not greater, and there would have been further indirect costs, such as court and judges' time and delays caused to other proceedings.  As a consequence the total costs of the Supreme Court decision are estimated at around £150,000 (176,000 euros).(JUSTICE and Maria Fletcher).

[105] The appeal process involved the refusal of Scottish courts to give permission to appeal and refusal to permit appeal to the Supreme Court. Cadder's representatives eventually succeeded in obtaining special leave to appeal to the Supreme Court under the Scotland Act 1998. Peter Cadder's legal costs throughout this process were borne by legal aid and have been estimated as at least £30,000 (35,000 euros).[105] The prosecution's costs falling directly on the state were likely to have been similar, if not greater, and there would have been further indirect costs, such as court and judges' time and delays caused to other proceedings.  As a consequence the total costs of the Supreme Court decision are estimated at around £150,000 (176,000 euros).(JUSTICE and Maria Fletcher).

[106] In the case of Malta and Ireland, not data were available so we have assumed that the cost was equivalent to that of the MS whose GDP per capita at PPP is the closet, namely Sweden in he case of Ireland and Portugal in the case of Malta.

[107] The appeal process involved the refusal of Scottish courts to give permission to appeal and refusal to permit appeal to the Supreme Court. Cadder's representatives eventually succeeded in obtaining special leave to appeal to the Supreme Court under the Scotland Act 1998. Peter Cadder's legal costs throughout this process were borne by legal aid and have been estimated as at least £30,000 (35,000 euros).[107] The prosecution's costs falling directly on the state were likely to have been similar, if not greater, and there would have been further indirect costs, such as court and judges' time and delays caused to other proceedings.  As a consequence the total costs of the Supreme Court decision are estimated at around £150,000 (176,000 euros).(JUSTICE and Maria Fletcher).

[108] Framework Decision 2002/584/JHA of 13 June 2002 on the European arrest warrant and the surrender procedures between Member States OJ L190/1 18.07.2002.

[109] ECJ Case C-396/11.

[110] Court of Justice of the European Union, (Grand Chamber), 26 February 2013, case C-399/11.

[111] Council Framework Decision 2009/299/JHA of 26 February 2009 amending Framework Decisions 2002/584/JHA, 2005/214/JHA, 2006/783/JHA, 2008/909/JHA and 2008/909/JHA, thereby enhancing the procedural rights of persons and fostering the application of the principle of mutual recognition to decisions rendered in the absence of the person concerned at the trial OJ L81/54 27.3.2009.

[112] ECJ Case C-168/13

[113] ECJ Case C-388/08

[114] ECJ Case C-192/12

[115] ECJ Case C-303/05

[116] Assange (Appellant) v The Swedish Prosecution Authority (Respondent) [2011] UKSC 22 On appeal from [2012] EWHC Admin 2849, 30 May 2012, Supreme Court of the United Kingdom

[117] Oberlandsgericht München, 16 Mai 2011.

[118] R (Gary Mann) v City of Westminster Magistrates’ Court & Anor [2010] EWHC 48 (Admin), Garry Norman MANN against Portugual and the United Kingdom 1 February 2011, Application no. 360/10, European Court of Human Rights (Fourth Section).

[119] Sofia City Court v Dimintrinka Atanasova-Kalaidzheiva, 9 September 2011, EWHC 2335.