TABLE OF CONTENTS

1........... Introduction.................................................................................................................... 2

2........... Procedural issues and consultation of interested parties.................................................... 2

2.1........ Chronology of the Impact Assessment and results of consultations................................... 2

2.1.1..... Policy context................................................................................................................. 2

2.1.2..... Organisation and timing................................................................................................... 2

2.1.3..... Consultation of interested parties..................................................................................... 2

2.2........ Consultation of the Impact Assessment Board................................................................. 2

3........... The recovery of criminal assets explained........................................................................ 2

3.1........ The asset recovery process............................................................................................. 2

3.2........ Fundamental questions of the asset confiscation and recovery process.............................. 2

3.3........ Asset recovery in the Member States and existing EU legal framework............................ 2

4........... Problem definition........................................................................................................... 2

4.1........ Insufficient recovery of criminal assets in the EU.............................................................. 2

4.2........ Components of the problem............................................................................................ 2

4.2.1..... Inadequate powers for confiscating criminal assets........................................................... 2

4.2.2..... Inadequate powers for preserving criminal assets pending confiscation............................. 2

4.2.3..... Inadequate powers for enforcing confiscation orders between Member States (mutual recognition)       2

4.2.4..... Powers for confiscation, preservation and enforcement are underutilised........................... 2

4.2.5..... Member States lack tools to maximise social utility from recovered assets (Redistribution) 2

4.3........ How would the problem evolve, all things being equal?.................................................... 2

4.4........ The EU's right to act and subsidiarity............................................................................... 2

4.4.1..... Conferral of power......................................................................................................... 2

4.4.2..... Legal basis .................................................................................................................... 2

4.5........ Fundamental rights.......................................................................................................... 2

5........... Policy objectives............................................................................................................. 2

6........... Policy options................................................................................................................. 2

6.1........ Identification and screening of "policy actions"................................................................. 2

6.2........ Discarded policy actions................................................................................................. 2

6.3........ Policy actions grouped into policy options....................................................................... 2

6.3.1..... Policy option 1 – Status quo........................................................................................... 2

6.3.2..... Policy option 2 – Non-legislative option.......................................................................... 2

6.3.3..... Policy option 3 – Minimal legislative option...................................................................... 2

6.3.4..... Policy option 4.1 – Maximal legislative option without mutual recognition......................... 2

6.3.5..... Policy option 4.2 – Maximal legislative option including mutual recognition....................... 2

7........... Analysis of the impacts of the policy options.................................................................... 2

7.1........ Analysis of Option 1 - Status Quo................................................................................... 2

7.2........ Analysis of Option 2 – Non-legislative option.................................................................. 2

7.3........ Analysis of Option 3 – Minimal legislative option............................................................. 2

7.4........ Analysis of Option 4.1 - Maximal legislative option without mutual recognition.................. 2

7.5........ Analysis of Option 4.2 - Maximal legislative option including mutual recognition................ 2

8........... Comparing the policy options.......................................................................................... 2

8.1........ Comparison of options and justification for choosing the preferred option......................... 2

8.2........ Proportionality and subsidiarity of the preferred option..................................................... 2

9........... Implementation, monitoring and Evaluation...................................................................... 2

Annex 1. Summary of fieldwork of the external IA study............................................................. 47

Annex 2. Confiscation statistics................................................................................................... 49

Annex 3. List of EU-level actions.................................................................................................. 2

Annex 4. Asset recovery in the UK............................................................................................. 58

Annex 5. EU27 profitability model.............................................................................................. 72

COMMISSION STAFF WORKING PAPER

Accompanying document to the Proposal for a DIRECTIVE OF THE EUROPEAN PARLIAMENT AND THE COUNCIL on the freezing and confiscation of proceeds of crime in the European Union IMPACT ASSESSMENT

Disclaimer: This report commits only the European Commission’s services involved in its preparation and does not prejudge the final form of any decision to be taken by the Commission

1.  Introduction

In order to disrupt organised crime activities it is essential to deprive criminals of the proceeds of crime. Organised crime groups are building large-scale international networks and amass substantial profits from various criminal activities. The proceeds of crime are laundered and re-injected into the legal economy.

The confiscation and recovery of criminal assets is considered as a very effective way to fight organised crime, which is essentially profit-driven. Seizing back as much of these profits as possible aims at hampering activities of criminal organisations, deterring criminality and providing additional funds to invest back into law enforcement activities or other crime prevention initiatives.

2. Procedural issues and consultation of interested parties 2.1. Chronology of the Impact Assessment and results of consultations 2.1.1. Policy context

As an effective tool, especially in the fight against organised crime, confiscation has been given strategic priority at EU level. The 2009 Stockholm Programme[1] highlights the importance of identifying and seizing criminal assets more effectively and re-using them where appropriate.

The Justice and Home Affairs Council Conclusions on confiscation and asset recovery adopted in June 2010[2] aim at promoting a more coordinated approach between Member States and achieve a more effective and widespread confiscation and recovery of criminal assets. They call on the Commission to consider strengthening the legal framework in order to achieve more effective regimes for third party confiscation and extended confiscation. They also highlight that attention should be focused on all phases of the confiscation and asset recovery process and recommend measures to ensure the preservation of assets during the confiscation process and the reuse of confiscated assets.

The Commission Communication "An Internal Security Strategy in Action"[3] indicates that the Commission will propose legislation in 2011 to strengthen the EU legal framework on confiscation, in particular to allow more third party confiscation and extended confiscation and to facilitate mutual recognition of non-conviction based confiscation orders between Member States.

The Commission Work Programme 2011 includes the proposal for a Directive on the confiscation and recovery of criminal assets as a strategic initiative which forms part of a broader package on the "protection of the licit economy", an agenda to protect Europe's economy which is closely linked to the EU 2020 Strategy. This package includes anti-corruption initiatives adopted in June 2011[4] and a Communication on an EU anti-fraud strategy[5].  

The envisaged legislative proposal on the confiscation [correct terminology? Does not "recovery" include confiscation? Needs to be checked in the Directive as well. Indeed it does. When reference is made to the Directive, it is better to use "confiscation", as the Directive mostly covers the legal proceedings. The asset recovery process also includes the asset tracing phase (eg national financial investigations, the work of the AROs) and the disposal phase (eg sale of an asset in a public auction or reuse of the asset for public purposes).]of criminal assets is also in line with the ten strategic priorities emphasised by the Commission in its Communication on the proceeds of crime adopted in 2008[6]. This Communication highlights shortcomings in the EU legal framework (lack of implementation, lack of clarity of some provisions, lack of coherence between existing provisions) and proposes to amend it. It further states that a revision would also allow to introduce new provisions in order to achieve a more coherent and comprehensive framework.

This impact assessment serves as a basis for the above legal proposal.

2.1.2. Organisation and timing

Work on the impact assessment started in May 2010 with the launch of an external study to support the preparation of the Impact Assessment[7]. The identification and finalisation of problems, objectives, policy options and assessment of impacts presented in this report were informed by the study which was completed in March 2011.

The study is based on a broad consultation of practitioners and experts, including interviews with some national contact points of the Camden Asset Recovery Inter-agency Network (CARIN)[8], and a limited consultation of other stakeholders. The results of these consultations are discussed below. As shown by their position in negotiating the JHA Council Conclusions, Member States generally agree that more needs to be done on confiscation and asset recovery[9].

This impact assessment is also based on the conclusions and recommendations of another external study[10] [the study seems to have been removed from the link indicated in the footnote! Thanks ! We will ensure that both this study and the IA are posted online before the proposal is published.] entitled "Assessing the effectiveness of EU Member States' practices in the identification, tracing, freezing and confiscation of criminal assets", contracted by the Commission to Matrix Consultancy and finalised in 2009. This study analyses Member States' practices in confiscation, focusing in particular on what has proven effective at national level with a view to promoting and exchanging best practices. The study identified several obstacles to effective confiscation, such as conflicting legal traditions, resulting in the lack of a common approach to confiscation measures, difficulties in securing and maintaining assets, lack of resources and training, limited cross-agency contacts and a lack of a coherent and comparable statistical system.

Finally, the impact assessment is based on the implementation reports issued by the Commission on the existing EU legal acts (see infra, Section 3.3). The reports on Framework Decisions 2005/212/JHA[11], 2003/577/JHA[12] and 2006/783/JHA[13] show that Member States have been slow in transposing them and that the relevant provision have been often implemented in an incomplete or incorrect way. Only Council Decision 2007/845/JHA seems to have been implemented in a moderately satisfactory way[14].

Statistics on confiscation and asset recovery activities are scarce (Annex 2 contains most of the available data on assets recovered). Reliable data sources on the number of ongoing freezing and confiscation procedures (especially those to be executed in other Member States), the turnover of criminal organisations, the costs of judicial procedures or the administrative costs related to asset management or data collection activities are even scarcer. Therefore, the economic impacts of the foreseen actions are often difficult to quantify.

2.1.3. Consultation of interested parties

Wide consultations and discussions with experts were carried out in the CARIN Plenary meeting (September 2010) and in seven meetings of the EU informal Asset Recovery Offices' Platform between 2009 and 2011. An expert meeting with Member States was held in October 2011.

Confiscation and asset recovery issues are also widely discussed between experts. International practitioners' meetings[15] and strategic seminars on confiscation and asset recovery[16] are increasingly taking place. Practitioners consider that most of the policy actions foreseen are best practices and as such have been included in the recommendations issued by CARIN between 2005 and 2010.

While the governments of the Member States were not formally consulted, they expressed their position on these issues last year in the negotiations of the JHA Council Conclusions mentioned earlier. Although there was broad agreement on most issues, a few Member States held a more reserved position on non-conviction based confiscation (e.g. Romania due to constitutional issues). Conversely, on other issues (eg third party confiscation, asset management) Member States agreed that more should be done, but suggested different solutions due to differences in their legislation, structures and practices.

Defence lawyers expressed concerns about the increased use of extended confiscation, non-conviction based confiscation and third party confiscation due to fundamental rights concerns (possible limitation of the right to property and of the right to a fair trial).

No open Internet consultation was carried out, as confiscation is a rather specialised topic where limited expertise is available but contacts were established with civil society, notably with the organisations promoting legality, the fight against organised crime and the protection of the victims of crime.[17] These organisations fully supported the envisaged measures.

During recent hearings on organised crime in the European Parliament's LIBE Committee, several Members of Parliament expressed a keen interest in strengthening the EU legislation on confiscation, mainly by using the existing Italian legislation (probably the most far-reaching) as a model.

The internal consultations in the Commission were mainly carried out through the inter-service group (ISG) on confiscation and asset recovery. The DGs and services represented in this ISG were DG HOME, DG JUSTICE, OLAF, DG EEAS, the Secretariat General, the Legal Service, DG MARKT,  DG TAXUD and DG ECFIN. Three meetings of the ISG took place in 2011 (i.e. on 16 February, 14 April and 4 May 2011) before the submission of this impact assessment to the Impact Assessment Board. Another meeting of the ISG was held on 30 June 2011 to discuss the Impact Assessment Board's opinion. The ISG members were in principle supportive of the main issues addressed in the impact assessment and their comments and suggestions were considered during the drafting of this report.

In view of the above, minimum standards for consultation of interested parties have been met in the preparation of this impact assessment.

2.2. Consultation of the Impact Assessment Board

The Impact Assessment Board (IAB) reviewed a preliminary version of this impact assessment and delivered its opinion on 10 June 2011. The recommendations for improvement were accommodated in this revised version of the report. In particular, the following changes were made:

· The problem requiring EU intervention has been explained more precisely.

· The justification for the preferred option has been strengthened, in order to clarify why it can be considered proportionate despite fundamental rights concerns.

· Stakeholder views have been presented earlier and in more detail in the report and the limited consultation has been acknowledged.

· Some cost element likely to arise from implementation has been indicated.

· The objectives of the initiative have been clarified to enable a meaningful evaluation in the future.

3. The recovery of criminal assets explained 3.1. The asset recovery process

Confiscation applies in principle to all crimes (or at least to most criminal activities in some Member States). However, in practice it is more frequently applied to serious cases involving organised crime. Typical examples are crimes generating huge income and liquidity, such as drug trafficking. The proceeds of crime are then converted into assets ranging from cash held in bank accounts to real estate, vehicles, livestock, artworks, company shares, businesses, collector's items etc. State authorities should be able to expediently identify and trace such assets, freeze them and manage them properly once they have been frozen.

Confiscation and recovery of criminal assets are two stages of a legal process whereby criminal assets (proceeds or instrumentalities of crime) are recovered in favour of victims, deprived communities or the state. At the heart of this process lies the determination by a court that particular assets are criminal and, thereby, liable to confiscation. This typically takes the form of a confiscation order. The full process is illustrated in the table below.

Figure 1: Steps in the asset confiscation and recovery process

The first stage in the asset recovery process is the tracing and identification of assets. This phase involves law enforcement investigations (usually under the coordination of a prosecutor) and requires substantial financial investigation skills. National Asset Recovery Offices (AROs) play a key role in expediently providing information to other AROs on the assets located in their territory.

After criminal assets are located in one or more countries, judicial procedures are needed to first freeze them and later to confiscate them. Following their freezing [what does seizure imply? LS suggested deleting this wording in Directive. In some national legislations and international conventions freezing refers only to the money on bank account and real estate, while seizure refers to all other assets, including cash. The EU legislator referred only to freezing orders for any property in FD 2003/577/JHA, therefore we can delete "seizure" in the IA], assets should be properly managed between the time when they are frozen and the time when a confiscation order is issued, so that their value is maintained. After a confiscation order is issued by a court, its execution is carried out. In principle assets become the property of the executing Member State, which may sell them or re-use them as appropriate. 

Each stage in the asset recovery process poses different challenges. For example, AROs should have the necessary resources to operate effectively, including access to all relevant information. Prosecutors and judges should trust the legal system of other Member States in order to recognise and execute foreign freezing and confiscations orders. Authorities should have the necessary skills and expertise to manage different assets as well as handle procedures in cross-border cases. Attention should be focused on ensuring effectiveness in all phases of the confiscation and asset recovery process.

The challenges of the asset recovery process can therefore be summarised as follows:

how to identify criminal assets;

how to preserve these pending a confiscation order;

how to obtain a confiscation order so that they can be recovered; and

how to enforce these orders.

These questions represent stages of an attrition process as explained in figure 2.

Figure 2: Stages of attrition

3.2. Fundamental questions of the asset recovery process

The confiscation process is complicated in practice because sophisticated criminals attempt to conceal their illicit gains from investigators, taking whatever measures they can to put assets beyond the scope of confiscation laws or enforcement measures. In response to this, newer confiscation tools have been introduced, such as:

value confiscation, so that assets of equivalent value can be confiscated where specific criminal assets are outside the reach of investigators;

third party confiscation, so that assets can be confiscated from the third parties to whom they have been transferred; and

mutual recognition of confiscation orders, so that assets can be more efficiently confiscated from other jurisdictions.

These tools are designed to make it easier for authorities to recover criminal assets. However, the key question is what assets should be deemed ‘criminal’ and, thus, liable to confiscation. The traditional approach of ordinary confiscation is to confiscate assets linked to a specific crime, following a criminal conviction for that crime.

The availability of ordinary confiscation can never ensure the recovery of all criminal assets because authorities will not always be able to prove that assets are the proceeds of specific crimes. In some cases, a conviction will have been obtained for the relevant crime, but authorities will lack evidence that particular assets are in fact proceeds of this crime. In other cases, criminal assets will go unrecovered because there is no criminal conviction to serve as a basis for ordinary confiscation. Such cases consist of two types: i) those where authorities have sufficient evidence but a case cannot be brought because it is time-barred or because the defendant is too ill, has died or absconded, lacks legal capacity (e.g. is a minor or of unsound mind), or has immunity from prosecution or amnesty; and ii) those situations in which authorities have insufficient evidence to obtain a criminal conviction.  This typology of criminal assets is illustrated in Figure 3. Type 1 assets are those amenable to ordinary confiscation proceedings; type 2 assets are those not so amenable due to barriers to prosecution; type 3 assets are those not so amenable due to insufficient evidence; type 4 assets are those not so amenable for both of these reasons.

Figure 3: Typology of criminal assets

Yet here too there have been legal developments in favour of the state powers. Examples include:

extended confiscation, in which a criminal conviction is followed by the confiscation not only of assets associated with the specific crime, but of additional assets which the court determines are the proceeds of other, unspecified crimes;

non-conviction based confiscation, in which civil, administrative or criminal procedure applies to recover illicit assets; and

extended criminalisation, which involves defining non-traditional crimes, with the result that more assets are liable to confiscation.

The above demonstrates that the following three questions need to be addressed when designing the asset recovery process:

· what assets should be liable to confiscation (i.e. how to delineate between assets which are, and are not, ‘criminal’);

· how to confiscate and recover these assets; and

· what to do with the recovered assets.

The asset recovery process must also be designed in such a way that it meets fundamental rights concerns.

3.3. Asset recovery in the Member States and existing EU legal framework

Within the EU, each Member State’s asset confiscation laws have evolved organically in response to domestic imperatives and, more recently, an EU legal framework. By the time the EU began to act in 2001 some Member States had potent asset confiscation regimes, whilst others did not[18].

The current EU legal framework consists of four Council Framework Decisions (FD) and one Council Decision:

· Framework Decision 2001/500/JHA, which obliges Member States to enable confiscation, to allow value confiscation where the direct proceeds of crime cannot be seized and to ensure that requests from other Member States are treated with the same priority as domestic proceedings

· Framework Decision 2005/212/JHA, which harmonises confiscation laws. Ordinary confiscation, including value confiscation, must be available for all crimes punishable by 1 year imprisonment. Extended confiscation must be available for certain serious offences, when "committed within the framework of a criminal organisation";

· Framework Decision 2003/577/JHA, which requires mutual recognition of freezing orders for a long list of crimes punishable by 3 years imprisonment, or if the ‘dual criminality’ principle is satisfied (i.e. for any offence punishable in both countries); and

· Framework Decision 2006/783/JHA, which mirrors these provisions for the mutual recognition of confiscation orders.

· Council Decision 2007/845/JHA on the exchange of information and cooperation between Asset Recovery Offices.

Each of these instruments was passed unanimously by the Council, exercising broad powers under Title VI of the Treaty on European Union. The context was in each case the fight against organised crime but, with the exception of the provisions on extended confiscation, the EU legal framework is not limited to organised criminal activity. 

4. Problem definition 4.1. Insufficient recovery of criminal assets in the EU

The problem addressed by this impact assessment is the insufficient recovery of criminal assets in the EU. To date, estimates of money lost to organised crime in the Member States, as well as data on success in asset recovery, remain patchy. Some recent reports and unofficial sources estimate the annual proceeds from organised crime in some Member States as very high[19]. Even considering a fraction of the estimated amounts as more credible, the figures provide a striking contrast with the amounts recovered annually in the Union[20]. Although only some Member States maintain statistics on the amounts recovered annually from crime, at present the number of freezing and confiscation procedures in the EU and the amounts recovered from organised crime seem modest if compared to the estimated revenues of organised criminal groups.

In the UK an official estimate in 2006 put organised criminal revenue at £15bn per annum.[21] Meanwhile, the UK’s Joint Asset Recovery Database (JARD) recorded approximately £125m worth of recoveries that year (see Annex 6)[22].  In Italy, organised criminal revenues were estimated at €150bn per year and the costs of corruption at 50-60 bn per year[23], eclipsing the amounts recovered and returned annually to the state (including to communities via social reuse programmes) (see Annex 2). Data from the UK and Italy therefore indicates that a low proportion of criminal assets are recovered. Moreover, comparing data on an annual basis (the value of assets recovered versus criminal turnover) is in fact a conservative approach, as it ignores unrecovered amounts from previous years which remain recoverable. 

Although reliable data sources are indeed scarce[24], the value of assets recovered in the EU can be considered insufficient, especially if compared to the estimated revenues of organised criminal groups or to the number of criminal convictions decided by courts for serious crimes.

Organised crime activities are often transnational in nature and the assets of criminal groups are increasingly invested in other Member States[25]. A good example is the recent operation "Shovel", described in Box 1, which shows the ramification of criminal activities and that important means need to be deployed to fight them successfully.

Box 1 – Operation "Shovel": The operation "Shovel" (2010) was conducted by the Spanish authorities in collaboration with the UK, Ireland and Belgium and with the assistance of Europol. The targeted criminal group led by Irish and UK criminals was involved in drug and weapons trafficking, money laundering, forgery of documents and murders. Over 700 police officers were deployed in many Member States on the day of the operation (more than 145 persons and 100 companies were under control). "Shovel" resulted in 38 arrests (24 in Spain including two lawyers who facilitated money laundering operations, 12 in the UK, 1 in Ireland and 1 in Bulgaria) and in the freezing of 60 luxury properties on the ”Costa del Sol” and 25 Luxury cars. 180 bank accounts were also frozen.

Pursuing assets located abroad is invariably more problematic, due to the increased difficulties in tracing them and to legal obstacles in obtaining evidence and executing freezing or confiscation orders. Table 1 lists examples of assets confiscated in cross-border cases with the assistance of Eurojust in 2010.

Table 1: Examples of asset confiscation in cross-border cases

Member State || Assets subject to confiscation with Eurojust's assistance

Belgium || Two boats to be confiscated in Spain.

Bulgaria || €37,000 cash and real estate property confiscated in the UK

Germany || €100,000,000 confiscated in a large tax fraud case involving coordinated searches in 15 countries

Ireland || Substantial amounts of property confiscated in Spain and Ireland

Spain || Five cases involving confiscation of €112,000,000, €17,000,000, €1,000,000, €23,000, and €9,000,000

Italy || Provisional confiscation in the Netherlands of €400,000. Freezing of 800 kg of counterfeit products in 10 countries. Freezing of a luxury watch in Germany. Freezing of 300 kg of cocaine in Belgium, Spain, Italy and Czech Republic. Freezing of documents related to the registration of 100 vehicles in Germany. Freezing of 700 kg of hashish, one pc, mobile telephones and documents in France, Spain and the UK. Freezing of one server in Austria.

France || Property and vehicles in Italy, one ship and the freezing of 1,400 kg of cocaine.

Sweden || €1,685,800 confiscated in Sweden and a ship in another country

UK || All property and money of a main suspect (including a house, a speedboat and money with a total value in excess of €1,200,000). Several luxury vehicles in Spain.

Moreover, the penetration of organised crime into the licit economy, even if it takes place in a single Member State, affects the functioning of the whole EU Internal Market, not only of that country.  Even when managing licit businesses, organised crime groups often support these activities with the recourse to intimidation and corruption, thus altering competition and the smooth functioning of the Internal Market. The resulting loss of revenues affects both national and EU financial interests, even when it takes place in only one Member State.

The aims of asset recovery are realised not only when criminals are deprived of their ill-gotten gains, but when these are redistributed effectively. In particular, the impact of asset confiscation upon public confidence in the criminal justice system may be enhanced through redistribution and restorative justice.  [reuse wording. Yes, but not advocating reuse for social nor public purposes. ]

Although asset recovery is a popular concept with a basis in international law[26], EU law and Member State laws, these laws remain underdeveloped and underutilised. It is unlikely that any Member State confiscates a significant proportion of criminal assets and, accordingly, it is unlikely that the laws themselves are achieving their stated aim.

4.2. Components of the problem

There are essentially three problems in relation to the EU legislative framework: its incomplete or late transposition, the existence of diverging national provisions and the low utilisation of confiscation in practice. The most plausible underlying reason for the late or incomplete transposition of the existing legislation is the workload of the responsible national authorities. This seems demonstrated by the somewhat surprising fact that transposition of the EU acquis has been slow or partial also in countries (eg Italy and the UK) where fighting organised crime through confiscation is a well-established priority at national level and where national legislation is well developed. For example the Commission implementation report on the EU rules on extended confiscation (FD 2005/212/JHA) showed that most Member States are slow in putting in place measures to allow more widespread confiscation, with only 16 of them transposing in full or partially by end 2007. The situation has slightly improved since the report, but transposition is not yet complete today

However, this is not only a case of incomplete transposition. The lack of clarity of some EU provisions has resulted in a different (and often diverging) transposition into national law, further widening the differences in the national legislations. One example is the notion of extended confiscation, which requires Member States to choose between three alternative criteria for extended confiscation, or to adopt two or all three of them cumulatively. As a result, the mutual recognition of orders based on extended confiscation is problematic. Due to the lack of coherence between provisions in the 2005 and 2006 Framework Decisions, a Member State can refuse to execute an order issued in another Member State if the two countries have not chosen the same criteria. Moreover, as highlighted in the Commission implementation reports, Member States have often added conditions (for example additional grounds for refusing the mutual recognition of confiscation orders) which further limit the effectiveness of the EU provisions.

These problems are compounded by a low implementation in practice (under-utilisation) of confiscation, as evidenced by the gap between the estimated size of criminal profits in a country and the amounts confiscated, or by the gap between criminal convictions and number of cases when they have been followed up with effective confiscation. The inadequate implementation in practice of the EU legislation may be due to different reasons depending on the stakeholders involved. For example, practitioners requesting the execution of orders in other Member States find the ad hoc request forms unclear and difficult to use. Law enforcement officers and prosecutors may perceive asset tracing and confiscation as a drain of scarce and expensive resources (financial investigators) that does not always entail tangible results (hence the importance to inform widely on the profitability of confiscation). Among the judges there may be some "cultural" resistance to apply confiscation widely, as it is perceived as an additional punishment on an already convicted person. Bearing in mind the challenges encountered during the different stages of the asset recovery process, the problem of the insufficient recovery of criminal assets in the EU can be broken down into components. These are discussed in more detail below.

4.2.1. Inadequate powers for confiscating criminal assets

The national confiscation systems in the Member States differ substantially. As organised criminal groups operate without borders and increasingly acquire assets in other Member States, there is greater need for both harmonised substantive law provisions and effective mutual recognition procedures in order to enforce freezing and confiscation orders in other Member States.

The EU legislation described above harmonises only some provisions on the national confiscation systems. For example, while it is common practice for suspected or investigated persons to transfer their assets to a knowing third party with a view to avoid confiscation, there are no provisions at EU level on third party confiscation. Also, the current EU legal framework only applies to criminal proceedings and the issuance of confiscation orders generally requires a criminal conviction. On the other hand, some Member States apply also non-conviction based systems to deprive criminals of illicit profits, which are not provided for at EU level.

The limited EU legal framework in place presents shortcomings. As it consists of a series of measures proposed by Member States over time, it is not sufficiently coherent and consistent. It ultimately provides relatively little harmonisation and offers wide discretionality to the Member States in its implementation. Moreover, some provisions have been implemented poorly or in a diverging way.  

There are also indications (eg from expert reports within the framework of technical assistance projects financed by the Union or by Member States) that legal inadequacies and political/structural problems in a few Member States may be so significant that they largely prevent the use of confiscation and asset recovery against high-ranking organised criminals. While these shortcomings mainly relate to extended confiscation (which exists in EU legislation since 2005), even obtaining ordinary confiscation seems somewhat problematic in these countries, as demonstrated by the low number of confiscation cases and the modest amounts recovered every year.

For the reasons above, a complete and correct implementation of the existing legal framework would not be sufficient to address the problem of the insufficient recovery of criminal assets in the Union. Many national provisions are not harmonised at EU level and the Member States exercise their powers in a very different way. A summary description of the provisions and the shortcomings existing in each Member State, as well as the potential impact of the proposed measures is provided in Table 5 (infra).

Box 2 – Examples of differences in the Member States' national provisions On the scope of the targeted assets, Member States currently employ varying definitions of criminal "proceeds" (as this term is currently undefined within the EU legal framework). As a result, the indirect proceeds of crime cannot always be confiscated. On the question whether particular assets are proceeds, most Member States employ a criminal standard of proof (high in some countries, such as Germany) while others, such as the United Kingdom, use a civil standard of proof which facilitates confiscation. On the timing for the confiscation procedure, in many Member States, the opportunity to confiscate criminal assets ends when criminal proceedings are finalised. This encourages criminals to try to conceal assets for the duration of the criminal proceedings, so that the assets which "resurface" after its conclusions cannot be confiscated. This can also cause authorities to rush financial investigations (with the risk of missing some assets) in order to conform to timetables imposed by criminal procedure. On the existing barriers to prosecution, the EU legal framework contains no mandatory provision for the confiscation of assets which cannot be confiscated because criminal proceedings are not allowed to be brought despite sufficient evidence. The assets believed to be criminal even though there is insufficient evidence to obtain the criminal conviction are covered by Framework Decision 2005/212/JHA in a limited and complicated way. Another relevant barrier is that in some Member States it is impossible to confiscate criminal assets where a conviction cannot be obtained (eg because the suspect has died, fled the jurisdiction, is unable to stand trial due to mental illness, has immunity from prosecution, etc). Finally, confiscation from third-parties is an entirely optional aspect of the existing EU legal framework. Some national confiscation regimes do not apply to assets which have been passed on to third parties. Others do, but require proof of mala fides even where the third party is a relative or close associate who has received for less than market value.

 

These differences in legislation often become barriers to the mutual recognition of freezing and confiscation orders between Member States. It is widely recognised that a minimum level of harmonisation should exist in order to facilitate mutual recognition. An increased level of harmonisation is therefore needed in this area, in order to ensure in each Member State a minimum level of protection from criminal infiltration in the legal economy (through the acquisition of assets) and facilitating the mutual recognition of freezing and confiscation orders in other Member States.

4.2.2. Inadequate powers for preserving criminal assets pending confiscation

The EU legislation does not contain harmonised provisions on preservation of assets and does not cover the management of seized assets pending the confiscation procedure, which is addressed exclusively by national provisions.

On asset preservation, while all confiscation regimes are supported by freezing powers, in a minority of cases these do not apply to all assets liable to confiscation (eg not to assets in the possession of third parties or assets representing equivalent value). Moreover, Member States do not always have in place appropriate mechanisms to ensure that assets in danger of being hidden or transferred out of the jurisdiction are able to be immediately frozen/seized while the request for judicial freezing is pending.

On asset management, in many Member States, assets are managed by agents (court officials, prosecutors, and police) involved in the criminal proceeding. In some cases, they lack even basic powers to realise seized assets which are liable to decline in value (even where requested to do so by the affected person). More generally, while a few Member States have established dedicated centralised structures, in most Member States expertise in managing complex assets is lacking and the management of assets is not centralised. This shortcoming directly affects the stages of attrition described in Figure 2. In some cases the spread between the value of the assets frozen or seized in view of confiscation and the value of the assets recovered at the end of the confiscation procedure is quite significant.

Box 3 – Examples on the risks in managing different assets While cash and financial products do not create particular management issues, the management of perishable goods or of vehicles has provoked in some instances substantial loss of value. Even managing real estate can be problematic. Assets frozen when the real estate market is booming may lose substantial value if they are auctioned at the moment when the market has declined. Other items which could risk losing value between the freezing and the confiscation are artwork and livestock (eg race horses).

 

4.2.3. Inadequate powers for enforcing confiscation orders between Member States (mutual recognition)

The enforcement of freezing and confiscation orders is generally not problematic in a national context. Problems arise more often when enforcing in a Member State an order issued by a judge in another Member State. The existing EU legal framework seeks to address the cross-border aspects of confiscation through provisions on minimum harmonisation and mutual recognition. However, at present the EU legal framework is insufficient, not completely transposed, not correctly transposed and in a few cases lacks coherence[27].

Box 4 Examples of problems with the mutual recognition of freezing and confiscation orders between Member States Framework Decisions (FD) 2003/577/JHA (freezing orders) and 2006/783/JHA (confiscation orders) establish a mechanism whereby the judicial authority issuing an order to be enforced in another Member State can send it directly to the judicial authority in that country which is competent to execute it, by filling in a specific certificate. This mechanism derogates from mutual legal assistance procedures, under which such orders should be sent to Central Authorities in each Member State. However, the full benefits of the swifter procedure provided in the FDs may not be fully reaped. The lack of transposition or the partial transposition by some Member States of the existing mutual recognition obligations significantly hampers the enforcement of freezing and confiscation orders in other Member States. The report issued by the Commission on FD 2003/577/JHA shows significant delays in the transposition, with only 19 Member States transposing this FD fully or partially by December 2008. Even today transposition is still not satisfactory, with four Member States not having transposed. The implementation report issued by the Commission on FD 2006/783/JHA shows that only 13 Member States transposed this FD fully or partially. Again, today the situation has marginally improved but is not yet satisfactory. Regrettably, the legal framework is not only incompletely transposed. Its provisions are sometimes transposed into national law in a diverging way. FD 2003/577/JHA emphasises that the national legislations show numerous omissions or misinterpretations. The report on FD 2006/783/JHA highlights that almost all Member States included additional grounds to refuse the mutual recognition of confiscation orders issued in other Member States. Again, a complete and correct implementation of the existing legal framework would not be sufficient, as some EU legal provisions are not well coordinated. Footnote 27 describes the negative impact on mutual recognition of the three alternative criteria for extended confiscation in Framework Decision 2005/212/JHA and the provisions on the grounds for refusal of mutual recognition of confiscation orders laid down in Framework Decision 2006/783/JHA. The existing FDs on  mutual recognition are also limited in scope. Extended confiscation is not supported by FD 2003/577/JHA, whilst FD 2006/783/JHA supports extended confiscation only in a limited way. Neither requires mutual recognition of non-conviction based confiscation orders. These limitations handicap the ability of Member States to combat organised crime. This is especially true for non-conviction based confiscation, as the alternative mutual legal assistance route is relatively weak. The issues with the certificates which lead to under-utilisation of these procedures are described in paragraph 4.2.4. Finally, the existing legal framework includes two similar instruments, ie FD 2006/783/JHA on the mutual recognition of confiscation orders and FD 2005/214/JHA on the  mutual recognition of financial penalties, including compensation orders. This dichotomy between confiscation and compensation seems unnecessary given that the recovery mechanisms employed are very similar.

4.2.4. Powers for confiscation, preservation and enforcement are underutilised

Based on available statistics (Annex 2), there is a significant underutilisation of asset confiscation laws throughout the EU. This is in some cases a matter of law, but mostly due to other factors. The study on confiscation conducted by Matrix in 2009 shows how cultural differences in the Member States affect the general approach to confiscation. As a result some judges consider confiscation almost as an additional punishment of an already convicted person and are reluctant to apply it systematically.

In general the profitability of asset confiscation work is also poorly understood by government decision-makers in some Member States. This lack of understanding causes asset confiscation work to be viewed as a drain upon scarce resources. Also because of this factor, underutilisation of asset confiscation tools persists, notwithstanding ample rhetoric on the utility of confiscating and recovering criminal assets. At law enforcement level, in some Member States the tracing and identification of criminal assets are neglected in favour of the criminal investigation. This evidently hampers the possibility to confiscate criminal assets. Within the judicial system prosecutors have a discretionary power to request freezing and confiscation orders and the courts have a discretionary power to issue them. As a result they are underutilised. On a cross-border level, the requests for freezing are very often made alongside other requests (eg. a house search), so practitioners using the specific certificates in FDs 2003/577/JHA and 2006/783/JHA must complete additional paperwork (for mutual legal assistance mechanisms) and need to be familiar with many different instruments. Fieldwork revealed these to be significant barriers for practitioners. 

4.2.5. Member States tools  for maximising social utility from recovered assets (Redistribution)

The existing EU legislation does not contain provisions on the disposal of assets. National provisions exist in some countries, including on the reuse of confiscated assets (eg for social purposes). In practice not all Member States have redistribution schemes in place. The assets recovered from organised crime are often sold in public auctions and the proceeds are returned to the State budget. In some cases organised criminal groups have been able to re-acquire the confiscated assets by discouraging potential bidders through intimidation. In countries where a system of decreasing value auction is in place (ie the price of the auctioned asset is progressively reduced if no one bids for it) they have also been able to pay a very low price. So this factor may also reduce State revenues resulting from recovered assets and affect the stages of attrition described in Figure 2. Moreover, the social reuse schemes established in some Member States take various forms. In some cases confiscated assets are directly put to social purposes, in others income streams are used to fund social benefits.

4.3. How would the problem evolve, all things being equal?

The baseline scenario or status quo indicates how the identified problem is likely to evolve without additional public intervention, taking account of existing and forthcoming interventions and following the introduction of the Lisbon Treaty.

Based on historical progress, at EU level the status quo policy option would likely result in an increase in the assets owned or controlled by criminal organisations, as well as an increase in their acquisitions of assets in other Member States. Evidence of a progressive increase in cross border criminality and in the cross-border acquisition of assets can be derived from investigative sources. The threat assessments issued by Europol[28] show an increasing trend in cross-border criminal activities and in the links between criminal groups located in different regions. The number of investigations coordinated by Europol, facilitated by Eurojust or supported by Joint Investigation Teams is steadily increasing. To some extent, the increasing cross-border acquisition of assets by organised crime is demonstrated by the steadily increasing number of requests for information on assets which are exchanged on a daily basis between Asset Recovery Offices in the Member States.

The expected increase in the cross-border acquisition of criminal assets would be mitigated to some extent by a corresponding slight increase of the amounts frozen or seized, of the amounts confiscated, of the amounts recovered and of the cases where mutual recognition of orders issued in other Member States is successful. Such increases would partly result from a slightly better transposition of the EU instruments. Better transposition could entail a wider recourse to extended confiscation and an increased number of cases of successful mutual recognition of freezing and confiscation orders issued in other Member States. The TFEU provides for the possibility of using infringements procedures to ensure transposition of the EU acquis in the area of Freedom, Security and Justice. However, even if these procedures may serve as an incentive for the Member States to act, their benefits will not be visible in the short-term.

Moreover, as the EU legislation currently harmonises only some provisions on the national confiscation systems, significant gaps would persist and would continue hampering a more successful recovery of criminal assets across the Union. For example, while it is common practice for criminals to transfer their assets to a knowing third party with a view to avoiding confiscation, there are no binding provisions at EU level on third party confiscation. The current EU legal framework only applies to criminal proceedings and the issuance of confiscation orders generally requires a criminal conviction. Non-conviction based rules to deprive criminals of illicit profits, which are successfully used in some Member States, are not provided for at EU level. While assets tend to decrease in value in the period between their freezing and the confiscation, the EU legislation does not contain harmonised provisions on preservation of assets and the management of seized assets pending the confiscation procedure.

At national level the progressive exchange of best practices and cross-border judicial cooperation between Member States will result in a slight increase of the utilisation of existing instruments, which should result in more successful asset investigations and confiscations. However, this progress in cross-border law enforcement and judicial cooperation may be partially offset by the expected increase in cross-border criminal activities, in the links between organised criminal groups and in the revenues of criminal organisations. On the other hand, the collection of statistics to measure the actual extent of confiscation and asset recovery activities (notably judicial statistics) and the corresponding costs will remain patchy, rendering any comparison between Member States difficult.

At international level the EU Member States will progressively sign and ratify the 2005 Council of Europe Convention (CETS 198). This new set of international obligations may induce some Member States to amend their national legislation to align it with the provisions of the Convention.While this Convention is based on a relatively good consensus, seven EU Member States have not even signed it yet. Without EU action, the implementation of the provisions of this Convention only by some Member States may therefore further widen the differences in the legislation between the Member States, at least in the short/medium term.

In any case action at EU level would still be necessary in order to address the shortcomings identified above. The shortcomings on asset preservation and reuse and on utilisation of confiscation powers are not addressed in the Convention. The existing EU legislation on the powers to confiscate (e.g. on extended confiscation) is more detailed than the provisions of the Convention. Finally, the enforcement of orders in other Member States is based on the principle of mutual recognition instead of mutual legal assistance.

As a result, the spread between criminally owned assets and assets recovered by governments is likely to increase and the cross-border dimension of confiscation is likely to gain relevance. We will be even more in a situation where "crime does pay". It is likely that the current costs of confiscation procedures would remain unchanged.

4.4. The EU's right to act and subsidiarity 4.4.1. Conferral of power

The EU has already passed measures relating to the confiscation and recovery of criminal assets. However, the EU’s conferral of power has changed following the entry into force of the Treaty of Lisbon. Whereas action under the old Third Pillar was essentially unconstrained provided all Member States agreed, the Treaty of Lisbon places specific conditions upon the EU’s right to act.

4.4.2. Legal basis

The legal basis to support action in the field of confiscation and recovery of assets can be found in Article 82(1) TFEU for the provisions on mutual recognition and in Article 82(2) and Article 83 (1) TFEU for harmonisation.

Under Article 5(3) TEU, the Union shall only act if the proposed action cannot be sufficiently achieved by the Member States. Article 67 TFEU foresees that the Union shall provide citizens with a high level of security by preventing and combating crime. Pursuing criminal assets is increasingly recognised as an essential tool to combat organised crime, which is very often transnational in nature and thus needs to be tackled on a common basis. This is all the more true in the EU, where the abolition of internal frontiers makes it far easier to commit cross-border crimes.

As acknowledged by the Stockholm Programme, the Union must reduce the number of opportunities available to organised crime as a result of a globalised economy, not least during a crisis that is exacerbating the vulnerability of the financial system. The EU is therefore better placed than individual Member States in sharpening more efficiently one of the most effective tools to fight organised crime groups.

The assets of organised criminal groups are increasingly invested outside their home country (often in several countries). This double cross-border dimension (of organised crime activities and their investments) further justifies pan-European action to target the assets of organised criminal groups.

While cross-border criminal and asset investigations may occur in several countries, prosecution and the judicial activities leading to confiscation normally take place in only one Member State. The resulting freezing and confiscation orders then need to be enforced in other Member States. Therefore, while criminal activities and investments are increasingly cross-border, confiscation procedures remain essentially national. However, their cross-border dimension is immediately evident in the enforcement of orders in other Member States.

Moreover, the penetration of organised crime into the licit economy even of a single Member State has an inherent cross-border dimension, as this affects the functioning of the whole EU Internal Market.

Apart from the issue of cross-border organised crime, the free movement of persons (11.3 million EU citizens live in a foreign Member State) and capitals within the Union entails a need for action at supra-national level in enforcing judicial decisions, including those on asset freezing and confiscation.

4.5. Fundamental rights

Inasmuch as it deprives the offender of his or her possession, confiscation interferes with the right to property enshrined in Article 17 of the EU Charter of Fundamental Rights. This right is not however absolute: it can legitimately be subject to restrictions when the legislator pursues a valid objective of general interest or the need to protect the rights and freedoms of others, such as the prevention of organised crime. There must nevertheless be reasonable proportionality between the policy behind the law and its effect upon the individual.

In the decisions of the European Court on Human Rights (ECtHR), the proportionality test mainly depends on the application of the confiscation order in the particular case under examination. The Court gives great weight to the procedural guarantees in place: a measure will be usually proportional if the individual had effective means to contest it.

Article 17 guarantees a right to own, use, dispose and bequeath lawfully acquired possessions. This wording seems to corroborate the possibility to confiscate the direct and indirect proceeds of crime, which by definition have been proven to have illicit origin. In fact under this Article it seems unclear whether a right to property exists where said property has illicit origin.

Article 47 of the Charter guarantees the right to an effective remedy and the right to a fair trial. Inasmuch as confiscation orders interfere with the right to property, affected parties must be able to challenge such orders under the conditions set by this Article.

1) Ordinary conviction-based confiscation is generally perceived to be a legitimate restriction to the right to property guaranteed by Article 17 of the Charter, if coupled with procedural guarantees to secure the right of affected individuals to an effective remedy and to a fair trial provided in Article 47. In certain circumstances, this right must be exercised together with the right to presumption of innocence of Article 48, the principles of legality and proportionality of criminal offences and penalties of Article 49 and with the right not to be tried or punished twice in criminal proceedings for the same criminal offence of Article 50 of the Charter.

2) Non-conviction based and extended confiscation regimes enable interferences with the right to property, in the meaning of Article 17 of the Charter, without the said property being linked to a specific criminal conviction. These measures are in principle harder to justify as necessary and proportionate restrictions to the right to property. Since these regimes do not relate to assets for which a criminal conviction has been obtained, they may raise issues with regard to the presumption of innocence guaranteed by Article 48(1) of the Charter and Article 6(2) ECHR. 

As with non-conviction based confiscation, extended confiscation may raise questions with regard to the presumption of innocence, as it is by definition a process which enables confiscation without an established link between the asset and a particular criminal conviction. Extended confiscation regimes may also raise concerns with regard to Articles 49 of the Charter and 7 ECHR which spell out the principle of legality, including the non-retroactivity of criminal law, and the prohibition against the imposition of harsher penalties. An issue may in particular arise in respect of newly-introduced extended confiscation provisions which allow for the confiscation of assets acquired through criminal conduct which occurred prior to the introduction of the extended confiscation regime.

3) The confiscation of assets transferred to third parties also affects the right to property within the meaning of Article 17 of the Charter. Again, the main issue is about proportionality, ie whether this limitation is proportionate to the objectives being pursued (deterring crime, restitution to victims).

4) Although freezing orders are only temporary measures, their consequences can be far reaching, particularly with regard to the right to property, despite criminal liability having yet to be established. On the other hand, they are necessary to ensure the subsequent application of confiscation orders, and have been upheld on this basis and justified because of public interest. Freezing orders can also raise issues with regard to the right to private and family life guaranteed by Article 7 of the Charter, and there is a growing awareness that to ensure compatibility with this right, as well as with the respect for human dignity in line with Article 1 of the Charter, Member States are expected to apply freezing orders in a way which leave people with their basic means of survival. Freezing orders can also have effects on third parties since it is often a criminal offence to have any dealings, commercial or otherwise, with a person on whom such a measure has been imposed.

5. Policy objectives

The overall long-term objective is the substantial reduction of organised crime revenues and accumulated wealth within the EU. The following general objectives for the confiscation and recovery of criminal assets, which align to general objectives of the Union in the Treaty of Lisbon, have been identified:

· to combat organised crime

· to achieve justice for victims

· to raise public confidence in the criminal justice system. 

The specific objectives of the EU intervention aim at both harmonising Member States practices in order to facilitate mutual recognition and to prompt asset recovey activities at national level in order to deter more effectively criminal activity. In line with the underlying components of the problem, four specific policy objectives can be defined:

· to further harmonise the confiscation powers of the Member States

· to harmonise the preservation powers of the Member States

· to enhance the enforcement of freezing and confiscation orders across Member States' borders

· to enhance the utilisation of existing tools in the Member States.

The table below shows the components of the problem, together with descriptions of the status quo at EU-level and associated specific and operational objectives.  It emphasises that the existing EU legal framework is far from comprehensive.

Table 2: Problem and objectives

Problem || Existing EU legal framework || General objectives || Specific objectives || Operational objectives

Not always possible to confiscate criminal assets due to gaps in MS powers (barriers to prosecution, insufficient evidence or both) || Rules are contained in FD 2005/212/JHA but many aspects of the problem are not addressed || 1/ combat crime (notably organised crime) 2/ achieve justice for victims 3/ raise public confidence in the criminal justice system || A. Increase the harmonisation of rules allowing to confiscate criminal assets, with due respect of fundamental rights || 1. harmonise confiscation of type 1 criminal assets (ordinary confiscation) 2. harmonise confiscation of type 2 criminal assets (barriers to prosecution) 3. harmonise confiscation of type 3 criminal assets (insufficient evidence) 4. harmonise confiscation of assets of third parties

Not always possible to freeze criminal assets, or preserve and manage frozen assets, due to gaps in MS preservation powers || No EU rules || B. Minimum harmonised rules allowing to freeze and manage criminal assets pending confiscation, with due respect of fundamental rights || 1. allow freezing orders for all assets liable to confiscation 2. To have effective mechanisms to preserve assets pending enforcement of freezing 3. To have effective systems for managing frozen/seized assets

Not always possible to enforce freezing and  confiscation orders across borders due to gaps in MS enforcement powers || FD 2003/577/JHA and FD 2006/783/JHA deal with mutual recognition of freezing and confiscation orders, but are limited in scope || C. Make it easier for MS to freeze and confiscate assets across borders || 1. MS to recognise and enforce (all types of) freezing orders from other MS 2. MS to recognise and enforce (all types of) confiscation orders from other MS

Underutilisation of freezing and confiscation procedures and tools by MS agents || No EU rules || D. Raise utilisation of freezing and confiscation tools by MS agents || 1. MS to raise utilisation of freezing powers 2. MS to raise utilisation of confiscation powers 3. MS to raise utilisation of mutual recognition instruments

|| || ||

6. Policy options 6.1. Identification and screening of "policy actions"

In order to meet the specific and operational objectives identified above and remedy the shortcomings resulting from the problem definition identified in Chapter 3. 21 EU-level policy actions were identified (some of which are complementary) which target particular operational objectives. These 21 policy actions are described in Annex 3, where they are grouped according to the four specific objectives they aim to achieve.

Given the high number of envisaged policy actions, the 21 EU-level actions were first screened individually against the following potential barriers to implementation: i) adequate conferral of power to the EU; ii) proportionality; iii) compatibility with fundamental constitutional or criminal law principles of the Member States.

The impact on fundamental rights was also analysed in detail, based on the relevant jurisprudence of the European Court of Human Rights. Whilst many of the identified policy actions affect fundamental rights, only in a small minority of cases it is not possible to remedy potential negative consequences. On the other hand, in some cases it appears that appropriate remedies can actually promote fundamental rights throughout the Union (by inducing a positive impact in Member States which currently afford low levels of protection).

6.2. Discarded policy actions

Four of the identified policy actions were discarded after the screening of the policy actions against the implementation barriers mentioned earlier:

· Civil standard of proof regarding whether an asset is "criminal" (policy action 3): The standard of proof on whether particular assets are proceeds could be harmonised to a lower "balance of probabilities" standard, to make it more difficult for convicted criminals to retain type 1 assets. This action was discarded due to likely problems with the conferral of powers to the EU and the proportionality principle, as well as problems of compatibility with Member States' legislation.

· Designating Asset Management Offices (policy action 11): Further harmonisation could require all Member States to entrust the management of frozen assets to Asset Management Offices at a national or regional level. This could increase efficiency and promote best practice. This action was discarded due to problems with the conferral of powers to the EU.

· Mandatory assets investigation (policy action 17): The EU legal framework could require investigators to open a parallel financial investigation, at least for the crimes listed in TFEU article 83(1).  This action was discarded due to problems with the conferral of powers to the EU and the proportionality principle and problems of compatibility with Member States' legislation.

· Limited judicial discretion (policy action 18): Judicial discretion could be limited by requiring freezing to be ordered wherever there is reasonable cause to suspect that an asset may become liable to confiscation and, in the event of a criminal conviction, by requiring confiscation to be ordered unless doing so would disproportionately affect fundamental rights. This action was discarded due to problems with the conferral of powers to the EU and the proportionality principle and problems of compatibility with Member States' legislation.

Doubts on the compatibility with the above implementation barriers could be expressed also in relation to other policy actions. However, such doubts were not so strong as to cause the relevant action to be discarded.

6.3. Policy actions grouped into policy options

Following the elimination of unfeasible policy actions, the remaining policy actions are grouped into policy options representing different degrees of EU-level intervention: a non-legislative option, a minimal legislative option (correcting deficiencies in the existing EU legal framework which inhibit it from functioning as intended) and a maximal legislative option (going beyond the aims of the existing EU legal framework). Within the latter, two maximal legislative sub-options are analysed, one with and one without EU level action relating to mutual recognition. The "do nothing" option forms the baseline against which all other options are analysed.

6.3.1. Policy option 1 – Status quo

This policy option would involve no new action at EU level, but constitutes the continuation of existing activities. The possible developments of this policy option at EU, national and international levels are indicated in section 4.3.

No new action at EU level does not mean no change at EU level. Protocol 36 to the Treaty of Lisbon ensures that the existing EU legal framework (or at least those provisions which do not exceed the EU’s post-Lisbon competence) will, on 1 December 2014, become enforceable against Member States through infringement proceedings brought by the Commission before the ECJ. Analysis will need to account for this step-change, as well as for other factors, including continued international developments and scrutiny in the forum of mutual evaluations by Moneyval and FATF.

6.3.2. Policy option 2 – Non-legislative option

Under the non-legislative policy option, workshops would be used to encourage Member States to better transpose the existing EU legal framework into domestic law (by highlighting its benefits and reiterating its compulsory nature) and to better utilise their asset confiscation laws (by highlighting benefits and sharing scientific knowledge and best practice). Better transposition can be achieved by promoting implementation of existing confiscation obligations (policy action 1). Although the trends towards compliance with FD 2005/212/JHA are positive, continued implementation/expert workshops could help ensure ongoing progress. Better utilisation of national legislation can be achieved by promoting implementation of existing mutual recognition obligations (policy action 12) via implementation/expert workshops on Framework Decisions 2003/577/JHA (freezing orders) and 2006/783/JHA (confiscation orders). Utilisation workshops for government decision-makers in some Member States on the profitability of asset confiscation work (policy action 15) could increase utilisation of these tools and provide a forum for the sharing of knowledge and practitioner experience.

6.3.3. Policy option 3 – Minimal legislative option

This option consists of transposition and utilisation workshops plus the policy actions aiming at consolidating confiscation and compensation orders (which concern the return of confiscated assets as compensation to identifiable victims of crime) and at providing consolidated mutual recognition forms. These additional policy actions deal with identified deficiencies in the existing legal framework on mutual recognition. 

In relation to the enforcement of confiscation orders, this option envisages mutual recognition of compensation orders (policy action 14). The legal framework could be simplified by consolidating FD 2006/783/JHA and 2005/214/JHA and extending their scope to include all compensation orders made in the context of criminal proceedings.

With regard to utilisation, this option would include the introduction of consolidated mutual recognition forms (policy action 19). A single form for all types of mutual recognition at the investigative stage could be provided within the European Investigation Order (and by suppressing the existing mutual legal assistance alternative). This option would also entail enforcing the primacy of mutual recognition (policy action 20). The EU legal framework could suppress the use of mutual legal assistance with respect to freezing and confiscation by repealing the existing mutual legal assistance conventions as regards requests between Member States.

6.3.4. Policy option 4.1 – Maximal legislative option without mutual recognition

This option would introduce many new aspects into the EU legal framework. It would consist of the transposition and utilisation workshops coupled with the policy actions aimed at further harmonising the confiscation, preservation and, to some extent, enhancing enforcement, i.e. all policy actions which do not involve legislative action in relation to mutual recognition.

In order to enhance confiscation powers, this option would foresee the possibility of confiscating all valuable benefits, including indirect proceeds (policy action 2). The EU legal framework could harmonise a (wide) definition of criminal ‘proceeds’ in order to ensure the recovery of ‘indirect’ proceeds resulting from the appreciation in value, or profitable reinvestment, of direct proceeds.  Harmonisation could also ensure that any valuable benefit (including, for example, the value of liabilities avoided) is liable to confiscation.

As a way to address the foreclosure of confiscation activities when the criminal procedure is concluded, this option foresees separating confiscation proceedings from criminal proceedings (policy action 4). The EU legal framework could ensure that separate confiscation proceedings can be brought also at a later date when criminal proceedings are finalised.

This option would also include strengthen extended confiscation (policy action 5) by providing for extended confiscation at least where a court finds it substantially more probable that the assets of a person convicted of an offence covered by Article 83(1) TFEU are derived from other similar criminal activities. 

With a view to addressing the identified barriers to prosecution, this option would include the introduction of non-conviction based confiscation in limited circumstances (policy action 6). The EU legal framework could make ordinary confiscation possible in circumstances where a conviction cannot be obtained because the suspect has died, fled prosecution or sentencing or is unable to stand trial due to permanent illness.

As criminals often transfer their assets to knowing third parties as soon as they are under investigation in order to avoid confiscation, this option would also include third party confiscation (policy action 7) in some cases. Laws could be harmonised by requiring third party confiscation to be available for assets received for less than market value and which a reasonable person in the position of the third party would suspect to be derived from crime

Policy actions 5, 6 and 7 have been conceived in a targeted way in order to comply with the principle of proportionality and take into account of the concerns expressed. Extended confiscation powers are already provided for in EU legislation (FD 2005/212/JHA). They already can be used only in case of serious crimes and have been applied in practice in a quite limited number of cases. Action 5 does not intend to enlarge the scope of extended confiscation. It merely intends to streamline the existing system of alternative criteria and options for Member States by providing a single minimum criterion for extended confiscation. The envisaged provision would not introduce a totally new obligation and would propose a minimum criterion (which is neither the lowest nor the most extreme) with which most Member States may be comfortable. If adopted, this provision would probably require only a few Member States to amend their legislation and bring it beyond their existing minimum.

In order to comply with the principle of proportionality, non-conviction based confiscation would not be introduced in all cases (full harmonisation), but would be allowed only in very limited circumstances where a criminal conviction cannot be obtained, eg because the suspect has died, fled the jurisdiction, or is unable to stand trial due to permanent illness. This provision has been also enshrined in the United Nation Convention against Corruption (Art. 54.1.c), which has been ratified by the Union and by 25 Member States.

Equally, under Action 7 third party confiscation is not foreseen in all cases, but only in limited circumstances (ie for assets which a reasonable person in the position of the third party would suspect to be derived from crime  and which have beenreceived for less than market value). This action would not affect the position of a bona fide third party who has acquired an asset paying its market value. Moreover, third party confiscation would take place only after an assesment, based on specific facts, showing that confiscation of assets directly from the person who transferred them is unlikely to succeed, or in situations where unique objects must be restored to their rightful owner.

In relation to the freezing/ of criminal assets, this option foresees the introduction of standards of universal freezing (policy action 8). Harmonised minimum standards could ensure that it is possible to preserve any assets and would facilitate the mutual recognition of freezing orders. It would also foresee mechanisms to safeguard freezing (policy action 9), so that  Member States would be required to have in place appropriate mechanisms to ensure that assets in danger of being hidden or transferred out of their jurisdiction can be frozen/seized immediately. This would include, in appropriate circumstances, the ability to freeze/seize prior to seeking a court order.

With regard to asset preservation, this option would grant powers to realise frozen assets (policy action 10). Harmonisation could ensure that, regardless of how frozen assets are managed, there are powers to realise them at least where they are liable to decline in value or uneconomical to maintain.

In the area of utilisation of powers, this option would introduce reporting obligations (policy action 16) for Member States, for example an obligation to report for all serious crimes covered by TFEU article 83(1), the assets frozen, the confiscation orders (if any) obtained and the type of order (eg ordinary, extended, non-conviction based confiscation). This would also help generate statistics which could be used for evaluation purposes.

Policy option 4.2 – Maximal legislative option including mutual recognition

This option consists of all the envisaged policy actions (but policy actions 19 and 20 partly overlap). Compared to option 4.1, this means that it also includes important provisions which foresee the mutual recognition of all types of orders (policy action 13). The EU legal framework could remove existing limitations on the mutual recognition of freezing and confiscation orders, allowing orders to better circulate around the Union.  This would also make the legal framework more coherent. This option would also entail the mutual recognition of compensation orders (policy action 14). The legal framework could be simplified by consolidating FD 2006/783/JHA and 2005/214/JHA and extending their scope to include all compensation orders made in the context of criminal proceedings.

As under the minimal legislative policy option, this option would also provide for consolidated mutual recognition forms (policy action 19) and for measures enforcing the primacy of mutual recognition (policy action 20).

7. Analysis of the impacts of the policy options

The five policy options (including two suboptions under maximal legislative option) have been assessed against an overall estimate of their social, economic and environmental impacts. No environmental impacts were identified.

Given the scarcity of measurable indicators and the lack of a coherent comparative statistics system on confiscation and asset recovery across the EU, it is almost impossible to quantify with precision the potential impacts of the policy options (some including policy actions which are very different in nature). Qualitative assessment therefore complements the analysis when quantification is not possible.

The figure below shows the main social and economic impacts and highlights in bold those which can be analysed meaningfully. Impacts flowing from a reduction in crime cannot be analysed because there is insufficient evidence that asset confiscation will reduce crime.

Figure 5: Impacts which can be assessed meaningfully

7.1. Analysis of Option 1 - Status Quo

For the reasons indicated in Section 3.5, on the ‘do nothing’ option, slow progress can be predicted towards achieving each of the specific objectives (i.e. confiscation, preservation, enforcement, utilisation and distribution).

Economic impact: Is expected to be low, resulting from natural progression in assets recovered. The spread between criminally owned assets and assets recovered by governments is likely to increase.

Social impact: Is likely to be negligible, mostly resulting from an increased application of social reuse practices across Europe and possibly leading to a slight increase in criminal assets being recovered in favour of crime victims.

Impact on criminal behaviour: Without additional action at EU level, criminals are likely to continue investing their assets in other Member States, thereby increasing the need for a cross-border dimension of confiscation activities.

Overall assessment of Option 1: The analysis in Section 4.3 revealed significant gaps, mutual recognition instruments would remain underutilised and the amount of criminal assets confiscated throughout the EU would remain small compared to estimates of organised criminal turnover. Whilst the situation without EU intervention would not be static, the pace of change would be too slow. This option would therefore not achieve the objective of increasing the recovery of criminal assets in the Union. Member States support for this option is unlikely. The European Parliament is expected to be totally dissatisfied with this solution.

7.2. Analysis of Option 2 – Non-legislative option

Economic impact: Transposition workshops on the EU legislation in force would not be particularly expensive and could entail a slight positive impact on transposition (and ultimately on assets recovered) by speeding up the process for the Member States which have not yet fully transposed the relevant texts. The costs of utilisation workshop would depend on the scale on which they are organised and their usefulness would likely be proportional to their scale. Given the severe underutilisation of confiscation procedures (as evidenced in the Matrix Study, see footnote 10) utilisation workshops could potentially have a more significant impact upon utilisation and avoid that decisions may continue to be made based on the assumption that asset confiscation work is unprofitable. Their added value would be proportional to the creation of a suitable evidence base. Transposition workshops on adopted EU legislation are regularly organised in Brussels. A meeting organised at the Commission premises, using the Commission interpreters and covering the travel expenses of two government experts per Member State costs around € 30 000. Utilisation workshops organised at EU level by the Commission would entail similar costs. If organised at national level, utilisation workshops are likely to be less expensive (lower travel costs, no interpretation needed) than those organised at EU level.

Social impact: The direct impact would be negligible, as workshops would impact on State authorities. A complete and correct transposition of EU legislation on extended confiscation and mutual recognition in the Member States and utilisation of confiscation tools would indirectly result in more compensation to victims and increased confidence in the criminal justice system. An increase in the use of extended confiscation would correspond to an increased limitation of  the fundamental rights (right to property, right to a fair trial, presumption of innocence) of the defendant. It should normally be balanced by adequate safeguards in national legislation (if Framework Decision 2005/212/JHA is correctly transposed).

Impact on criminal behaviour: To the extent that this option can be expected to generate increased utilisation of existing tools, any impact on criminal behaviour is expected to be quite limited. While it may cause occasional criminal capital flight, this option will likely oblige criminals to better hide their assets, for example by increasingly transferring them to a knowing third party before their conviction in a criminal court. It may therefore have a feeble negative impact on the illicit economy and on the economies of third countries[29].

Overall assessment of Option 2: Overall the added value of the non-legislative option is likely to be low. Although transposition remains incomplete, there is only slight scope for non-legislative action alone to add value in circumstances where the existing legal framework will become enforceable by 2014 in any event[30]. The organisation of EU transposition workshops could also affect practitioners' perception and facilitate enforcement. However, the most promising aspect of the non-legislative option is the utilisation workshops. This option would hardly achieve the objective of increasing the recovery of criminal assets in the Union.  It would also draw heavy criticism from the European Parliament.

7.3. Analysis of Option 3 – Minimal legislative option

Economic impact: the impacts of transposition workshops (Actions 1 and 12) and utilisation workshops (Action 15) are described above. Improving mutual recognition instruments by consolidating confiscation and compensation orders (policy action 14), providing consolidated mutual recognition forms (policy action 19) and/or enhancing the primacy of mutual recognition (policy action 20) would clearly increase the number of cross-border enforcement procedures and, to some extent, the value of the assets recovered. However, it is hard to assess the economic added value of even a significant increase in the utilisation of mutual recognition instruments. An increased utilisation in mutual recognition instruments would shift administrative costs from central authorities (in charge of mutual legal assistance) to (local) judicial authorities. As mutual recognition is less convoluted than mutual legal assistance, the administrative cost of handling requests from other Member States should in principle decrease. The extent of this slight decrease in direct costs would depend on the relative efficiency of the different parts of Member State bureaucracies. The time savings resulting from a wider use of mutual recognition (as opposed to mutual legal assistance) would allow faster cross-border execution and increase the chances of successful recovery by limiting the risks for asset dissipation. The envisaged consolidation of mutual recognition forms may require some training for the practitioners to use the new single mutual recognition form. These costs would be likely offset by the benefits in the form of increased value of assets recovered (resulting from an increased number of cross-border enforcement procedures).

Social impact: A moderate increase in the number and value of assets recovered should logically correspond to a moderate increase in compensation to victims. Better enforcement of cross-border procedures would likely result in increased confidence in the national criminal justice systems and in the EU Area of Justice, Freedom and Security. In relation to fundamental rights, the increase in the cross-border enforcement of orders will concern ordinary confiscation. Because this is the procedure with the least impact on fundamental rights, a low impact on the right to property is expected.

Impact on criminal behaviour: Slight, as it results from an increased utilisation of existing tools and increased mutual recognition of orders issued in other countries. As under option 2, it would oblige criminals to better hide their assets (eg using third parties). A better enforcement of cross-border procedures may have some displacement effects, resulting in a net capital flight of criminal money out of the EU. It may therefore have a slight negative impact on the illicit economy and on the economies of third countries.

Overall assessment of Option 3: Overall the added value of the minimal legislative option is likely to be moderate. In addition to the (low) added value of option 2, the increased enforcement of freezing and confiscation orders in other Member States resulting from better mutual recognition instruments would likely produce limited economic and social impacts. However, policy actions 19 and 20, aimed at facilitating the mutual recognition of orders, would significantly enhance utilisation of mutual recognition instruments although action 19 has additional benefits over action 20. Moreover, an enhanced utilisation of mutual recognition over mutual legal assistance would substantially reduce the time necessary to enforce freezing and confiscation orders in other Member States. This option would barely achieve the objective of increasing the recovery of criminal assets in the Union. It would likely not be considered as an adequate response to the problem by the European Parliament.

7.4. Analysis of Option 4.1 - Maximal legislative option without mutual recognition

The maximal legislative option (in its sub-option without mutual recognition) builds upon the non-legislative option by introducing a number of new aspects into the existing EU legal framework, i.e. all policy actions which do not involve legislative action in relation to mutual recognition.

Economic impact: In addition to the impacts of transposition workshops (Actions 1 and 12) and utilisation workshops (Action 15) described above, it can be estimated that most of the actions in this policy option, considered in isolation, would have at least a moderately positive economic impact. For example, although policy action 10 on new powers to realise frozen assets would entail implementation costs (introduction of procedures to sell frozen assets), these would be largely or entirely offset by the benefits in the form of reduced ongoing costs for asset management and no decrease in the value of the assets. Policy Action 7 introducing confiscation from third parties would produce, even in its most limited form[31], at least a moderate increase in the confiscation powers and hence in the value of assets recovered. The application of policy action 6 introducing non-conviction based confiscation, even in limited circumstances has demonstrated that it may have a substantial impact on organised crime as illustrated in box 5 .

Box 5 – Operation "Nemesi": In Italy the application of non-conviction based confiscation provisions to a dead suspect's heirs has allowed in 2010 to freeze, in a single case, assets estimated at € 700 million at least. Dante Passarelli, a businessman suspected of being the "fiduciary person" of the head of the Camorra Casalesi clan, died in unexplained circumstances. He had been convicted of participation in a criminal organisation by a first instance criminal court, but an appeal was pending. The assets frozen (registered in his name or attributed to him by investigators) included 136 apartments, 11 warehouses, 75 land estates, 8 shops, 2 villas, 51 garages, company shares and bank accounts, for a total amount estimated between € 700 million and € 2 billion. In 2008 Italy passed legislation which could prevent the heirs from a deceased defendant, whose assets have been frozen or seized, from legally inheriting the assets and having them released. Mr Passarelli's wife and 6 children were not able to explain the licit origin of all these assets, nor the huge disproportion between their declared revenues and the frozen assets.

The reuse of confiscated assets for social purposes may also have economic benefits, allowing NGOs to start business activities using confiscated assets[32] which normally become profitable over time. On the other hand, separating criminal proceedings from confiscation proceedings (Action 4) would likely result in slightly increased direct costs (due to additional procedures). These costs can likely be offset by the expected increase in the assets recovered, resulting from being able to identify and pursue criminal assets for a longer time, even when criminal proceedings are over. Because of the severe lack of data in relation to amounts frozen, confiscated and recovered, and in relation to the costs of carrying out confiscation-related activities, it is not possible to provide a quantification of the overall cost of this option. Moreover, in many cases implementation costs may differ depending on the characteristics of the Member States, for example for social reuse programmes (Action 21) and in some cases costs would depend on how Member States would implement an action. For example, the separation of criminal and confiscation proceedings (Action 4) could be applied automatically in all cases or only in cases where this is considered necessary. The administrative burden related to the reporting obligations (Action 16) would also vary between Member States, depending on the extent to which they do or do not already collect some of the data required for reporting purposes.

In order to address the lack of data described earlier, the main economic analysis presented is an EU27 profitability estimate based on a model which uses proxy indicators to extrapolate from a detailed analysis of income and cost in the UK (details in Annexes 4 and 5). The UK is the only Member State for which income and costs for all elements of the asset confiscation system can be estimated. Its asset confiscation system is also a reasonable approximation of the maximal legislative sub-option under consideration. Although only indicative, the results of this exercise are encouraging: 21 of 27 Member States are indicated by the model to be profitable (many of them highly profitable) for the maximal legislative option in its sub-option without mutual recognition.

Table 3: Profitability of maximal legislative option without mutual recognition, EU27

Member State || Revenue (€m) || Cost (€m) || Profit (€m) || Profit ratio (profit/cost) || Categorisation

Czech Republic || 131.00 || 36.57 || 94.43 || 3.50 || Highly profitable

Lithuania || 188.08 || 131.08 || 57.00 || 2.63 || Highly profitable

Spain || 167.56 || 124.86 || 42.70 || 2.58 || Highly profitable

Latvia || 31.66 || 10.87 || 20.79 || 2.12 || Highly profitable

Poland || 19.21 || 4.27 || 14.94 || 1.91 || Highly profitable

Slovakia || 32.49 || 18.33 || 14.15 || 1.46 || Highly profitable

Slovenia || 105.42 || 91.92 || 13.49 || 1.37 || Highly profitable

Romania || 109.13 || 96.73 || 12.40 || 1.16 || Highly profitable

Estonia || 19.24 || 13.80 || 5.44 || 1.11 || Highly profitable

Bulgaria || 10.78 || 5.77 || 5.01 || 0.87 || moderately profitable

Hungary || 8.89 || 4.12 || 4.77 || 0.81 || moderately profitable

Netherlands || 12.88 || 8.32 || 4.56 || 0.77 || moderately profitable

Portugal || 8.02 || 4.43 || 3.59 || 0.55 || moderately profitable

Malta || 5.85 || 2.38 || 3.47 || 0.46 || moderately profitable

UK || 21.39 || 19.22 || 2.17 || 0.43 || moderately profitable

Cyprus || 2.74 || 0.75 || 1.99 || 0.42 || moderately profitable

Greece || 3.43 || 1.45 || 1.98 || 0.39 || moderately profitable

Italy || 2.37 || 0.76 || 1.61 || 0.34 || moderately profitable

France || 2.11 || 1.49 || 0.62 || 0.15 || moderately profitable

Germany || 1.00 || 0.48 || 0.53 || 0.13 || moderately profitable

Belgium || 0.26 || 0.18 || 0.08 || 0.11 || moderately profitable

Luxembourg || 1.48 || 1.51 || -0.03 || -0.02 || Not profitable

Sweden || 7.67 || 8.62 || -0.96 || -0.11 || Not profitable

Austria || 7.74 || 9.27 || -1.53 || -0.17 || Not profitable

Ireland || 7.66 || 9.89 || -2.23 || -0.23 || Not profitable

Finland || 1.99 || 4.33 || -2.35 || -0.54 || Not profitable

Denmark || 1.73 || 5.79 || -4.06 || -0.70 || Not profitable

The fact that asset confiscation work appears to be potentially profitable in most Member States pleads in favour of EU-level intervention For the minority of Member States for which asset confiscation may be unprofitable (mostly Nordic Member States where relatively low criminality and commensurately low investment in policing may results in less assets recovered) this does not detract from the case for EU-level intervention. In fact, even the maximal legislative options would oblige all Member States to transpose the new EU provisions into their legislation, but would not force reluctant Member States to incur the (higher) costs of increased utilisation of confiscation procedures.

(ii) Social impact: It can be assumed that recovering more assets in favour of the State will have a significant social impact as it will, provide funding for public authorities for provision of public services, including in favour of victims of crime.

(iii) Fundamental rights impacts: Actions having a significant positive impact upon confiscation tools (eg Action 5 on extended confiscation, 6 on non-conviction based confiscation or 7 on third party confiscation) are also those with the biggest impact on fundamental rights. Defence lawyers expressed concerns about their possible increased use. On the other hand, these measures were considered extremely important by investigators, prosecutors and other practitioners. A limitation of the right to property and right to a fair trial of the defendant must be justified, respect proportionality and be accompanied by adequate safeguards.

While conviction-based confiscation regimes, as such, are rarely problematic from a fundamental rights perspective, non-conviction based and extended confiscation regimes are more contentious. The ECtHR has rendered many decisions, consistently upholding their application in particular cases. However, it has avoided ruling on the principled question of their compatibility with the Convention. Since these regimes do not pursue solely a punitive objective, they have to be justified on broader grounds. With regard to Articles 47 (right to a fair trial) and 48 (presumption of innocence) of the Charter and to the corresponding Articles 6 and 13 of the European Convention on Human Rights (ECHR), non-conviction based confiscation regimes have consistently been held to be civil in nature, and the ECtHR has also refused to qualify extended confiscation as a ‘criminal charge’.

Reversals of the burden of proof concerning the legitimacy of assets have so far survived the scrutiny of the ECtHR, so long as they were applied fairly in the particular case, with adequate safeguards in place to allow the affected person to challenge these rebuttable  presumptions.  For example in a specific case the Italian regime was held to be a proportionate restriction in as much as it constitutes a “necessary weapon” in the fight against the Mafia. In another specific case the UK civil confiscation regime was upheld and considered more generally targeted at recovering criminal assets that did not lawfully belong to the applicant.

The right to be presumed innocent until proven guilty under Article 48 of the Charter only applies when a person has been charged under criminal law and not where the proceedings are civil in nature. Before the European Court of Human Rights, defendants in non-conviction based confiscation proceedings have argued that these are criminal and violate the right to the presumption of innocence, but these arguments have so far failed before the ECtHR.

It is also harder to justify the compatibility of extended confiscation with the right to property. Indeed, the confiscation order extends to assets beyond those derived from the offence for which the person was convicted in the main proceedings. It therefore has to be shown that it pursues a broader legitimate objective to that of punishing the individual. While the ECtHR has consistently upheld extended confiscation regimes in specific cases (eg the Italian regime, which is perhaps the most severe regime in the EU as it combines non-conviction based and extended confiscation), their compatibility with the Convention is assessed on a case by case basis. Again, the degree of procedural safeguards afforded to the defendant plays a determinant role in assessing the proportionality of the measure.  

A strong argument in favour of justifying third party confiscation is the case where assets are claimed both by the third party and by a victim. If the perpetrator of a crime has insufficient assets to meet a claim (as is often the case), measures in favour of the third party would weaken the position of the victim.

Temporary measures, such as freezing orders may, due to their provisional character, justify further limitations of certain rights and traditional principles of due process, provided sufficient safeguards or remedies are available and those limitations respect the essence of those rights and principles (compare Art. 52 of the Charter of Fundamental Rights). Many States use techniques such as ex parte or in camera proceedings to ensure that the affected person is not able to defeat the purpose of the order through prior knowledge of it. So far, the ECtHR has repeatedly held that the Italian procedural rules for ordering a preventative confiscation violate the right to a fair hearing as they do not foresee the possibility for the defendants to ask for a public hearing. It thus becomes even more pressing to avoid that national measures which may violate the ECHR or the Charter could benefit from EU rules on mutual recognition. 

If applied with proportionality and complemented with adequate safeguards, laid down in the EU legislative proposals, the measures in this policy option would respect fundamental rights.[33]. According to the jurisprudence of the ECtHR the existence of effective legal remedies is a pre-condition to ensure that fundamental rights are respected. Equally, under the Charter of Fundamental Rights, it is necessary that EU legislation itself contains sufficient procedural safeguards and remedies (see Box 7).

Box 7 – Legal remedies The existing EU legislation (eg Article 4 of Framework Decision 2005/212/JHA) provides that Member States should ensure that adequate legal remedies for the affected persons exist in national legislation. With a view to fully comply with the EU Charter on Fundamental Rights, safeguards are required at EU level in order to guarantee, the respect of the presumption of innocence, the right not to be tried or punished twice in criminal proceedings for the same criminal offence, the right to a fair trial, the existence of effective judicial remedies before a court and the right to be informed on how to exercise such remedies.

 (iv) Impact on criminal behaviour: Would be significant, as non-conviction based confiscation (even in limited circumstances) and third party confiscation would oblige criminals to change their practices and make it more difficult for them to hide their assets. This option could cause moderate capital flight of criminal money to non-EU countries and have a significant negative impact on the illicit economy and on the economies of third countries.

Overall assessment of Option 4.1: Overall the added value of the maximal legislative option (sub-option without mutual recognition) is likely to be significant. The fact that asset confiscation work appears to be potentially profitable in most Member States is a strong argument which significantly reduces the immediate need to demonstrate other benefits. However, significant social benefits can also be expected, provided that the actions most likely to affect fundamental rights are proportional in their scope and balanced by adequate safeguards. The immediate impacts of implementing this option include stronger systems for confiscation, freezing, managing and redistributing assets. However, this option would also bring an important impact on utilisation. Member States do not want to be seen to be performing poorly. While utilisation workshops would inform Member State decision-makers about the potential profitability of asset confiscation work and thus empower them to promote change, more powerful legislative tools would encourage utilisation by concretely raising the chances of successful intervention. Moreover, harmonisation of confiscation laws can also de facto promote mutual recognition by ensuring that incoming orders are compatible with the judicial system of the executing Member State. This option would achieve the objective of increasing the recovery of criminal assets in the Union. Most likely it would be moderately welcomed by the European Parliament.

7.5. Analysis of Option 4.2 - Maximal legislative option including mutual recognition

Economic impact: In economic terms, adding EU-level action on mutual recognition (policy actions 14, 19 and 20 aiming to ensure utilisation of mutual recognition instruments, the impacts of which are described in Section 7.3, and policy action 13 which aims at expanding the scope of mutual recognition to all orders) would improve the results of the EU27 profitability analysis still further. However, given the scarcity of data on the number and amounts of orders to be enforced in other Member States, a detailed profitability estimate by country for this policy option is not possible. The additional costs for Member States liable to receive many foreign non-conviction based orders for execution would be fully offset by the existing provision (in FD 2006/783/JHA) that the Member State enforcing a confiscation order is entitled to retain 50% of the recovered value[34].

Social impact: Significant. In addition to the impacts described in Section  7.4 (more assets recovered in favour of the victims and victimised communities and increased confidence in the national criminal justice systems, increased impact on fundamental rights which requires new provisions to comply with the principle of proportionality and be balanced by adequate safeguards) the important enhancements on mutual recognition can be expected to result in increased confidence in the EU Area of Justice, Freedom and Security.

Impact on criminal behaviour: Significant. The measures in the maximal legislative option (notably non-conviction based confiscation in limited circumstances and third party confiscation), coupled with a significantly improved enforcement of cross-border procedures (resulting especially from the expansion of mutual recognition to all types of orders, including non-conviction based orders) would likely oblige criminals to change their practices and could have displacement effects, resulting in a net capital flight of criminal money out of the EU. This would result in an even more significant impact on the illicit economy and on the economies of third countries.

Overall assessment of Option 4.2: Overall, the added value of the maximal legislative option (sub-option with mutual recognition) is likely to be very significant.  The combined effects of economic profitability, significant social impacts (both on victims and victimised communities through more assets recovered in favour of victims and more social reuse) and greater utilisation are further enhanced by actions on mutual recognition which are more far-reaching than those described in Section 7.3. Together with the other policy actions, the latter will significantly improve the status quo as regards cross-border enforcement of orders throughout the Union. This is important because barriers to enforcement are effectively a dampener on profitability, tending to discourage utilisation in Member States with non-conviction based confiscation regimes. This option would be fully consistent with the objective of increasing the recovery of criminal assets in the Union. It would likely be welcomed by the European Parliament.

8. Comparing the policy options 8.1. Comparison of options and justification for choosing the preferred option

As a reminder, the table below summarises the objectives, policy actions/options and their expected impacts. In the table ü indicates that there is no problem, ? indicates a potential problem, ?? indicates a likely problem, and û indicates a clear problem, in which case the action itself is struck out (screening). 

Impacts (applied vis-à-vis the ‘no change’ baseline) are rated + or – for slight impacts, ++ or – – for moderate impacts, and +++ or – – – for significant impacts. 

Table 4: comparison of policy actions and options against objectives and with expected impacts

The table below shows that the potential impacts of the policy actions in the Member States could be very different based on the differences in their legislation, structures and practices. There is a wide gap between countries which have only basic rules and structures in place, such as Greece, to countries with very sophisticated and effective systems, such as the Netherlands. In the table, ü denotes that the measure under consideration is already implemented, û denotes that it is not, P denotes partial implementation, A denotes an alternative approach to the objective, ? denotes a gap in the dataset and P/? denotes at least partial implementation.

Table 5 - Location of potential impacts by Member State

Bearing in mind the analysis of impacts, each policy option can be ranked in terms of its impact against the five specific objectives of enhancing confiscation, preservation, enforcement, utilisation and social utility from recovered assets (redistribution).  Based on the first four of these rankings, a ranking can also be derived for impact upon the number of assets recovered (which corresponds to the problem of insufficient recovery). In the tables below, rankings are expressed in decreasing order, 1 being the highest score.

Table 6: Preliminary ranking of options against specific objectives

Specific objective || Policy option rankings

No change || Non-legislative || Minimal legislative || maximal  without MR || Maximal with MR

Confiscation || 0 || 2 || 2 || 1 || 1

Preservation || 0 || 2 || 2 || 1 || 1

Enforcement || 0 || 3 || 2 || 3 || 1

Utilisation || 0 || 4 || 2 || 3 || 1

Redistribution || 0 || 4 || 3 || 2 || 1

Assets recovered* || 0 || 3 || 2 || 2 || 1

* Impact on number of assets recovered is a function of confiscation, preservation, enforcement and utilisation.

The maximal legislative options entail a higher impact on confiscation, preservation and redistribution with respect to the other options. The minimal legislative option impacts positively on enforcement and utilisation. The maximal legislative options entail a higher overall impact on assets recovered. The sub-option with mutual recognition (which includes all actions in the minimal legislative option and the maximal option without mutual recognition) would have the highest impact.

The table below shows the policy options ranked against each other with reference to the impacts, the potential barriers to implementation and other factors, using the same ranking criterion (1 being the highest). It finally shows an overall assessment in the form of a ranking.

Table 7: Ranking of options against impacts, barriers and criteria

Criteria || Policy option

no change || Non-legislative || minimal legislative || maximal  without MR || Maximal with MR

Economic Impacts || 0 || 4 || 3 || 2 || 1

Social Impacts || 0 || 4 || 3 || 2 || 1

Fundamental Rights || 0 || 1 || 1 || 2 || 2

Proportionality || 0 || 1 || 1 || 2 || 2

MS compatibility || 0 || 1 || 1 || 2 || 3

Simplicity & coherence || 0 || 4 || 2 || 3 || 1

Implementation costs || 0 || 1 || 1 || 2 || 2

Administrative burden || 0 || 1 || 1 || 2 || 2

Geographical disposition || 0 || 1 || 1 || 1 || 1

Overall assessment || 0 || 3 || 2 || 2 || 1

As described in Sections 7.4 and 7.5, the maximal legislative options entail higher economic and social impacts than the other options, while having a bigger impact also on fundamental rights, proportionality, compatibility, costs and administrative burden. This analysis clearly shows that the preferred policy option is the maximal option featuring action on mutual recognition. The maximal legislative option without action on mutual recognition and the minimal legislative option are ranked equal second.

8.2. Proportionality and subsidiarity of the preferred option

The preferred option respects the principles of subsidiarity and proportionality because it does not go beyond what is needed to achieve the objectives described in section 5 whilst respecting fundamental rights. The maximal legislative option including mutual recognition would considerably enhance the confiscation and enforcement powers of the Member States, inter alia by amending existing provisions on extended confiscation and introducing new provisions on non-conviction based confiscation, third party confiscation introducing the mutual recognition of all types of orders (including non-conviction based orders).

However, its policy actions would be calibrated in order to be proportionate and not unduly affect fundamental rights. For example the introduction of harmonised non-conviction based confiscation provisions is not foreseen in all cases, but only in very limited cases where the defendant cannot be prosecuted (due to death, illness, abscondence or immunity). Third party confiscation would not be allowed in all cases, but only when the acquiring third party should have suspected that the assets are proceeds of crime and paid less than market value. In order to meet the concerns expressed by defense lawyers, safeguards at EU level are foreseen with a view to fully comply with the EU Charter on Fundamental Rights. In order to reach an equivalent outcome, virtually all Member States would have to amend their national legislation. However, these legislative changes would not be coordinated and would in any case not address the mutual recognition of foreign orders (where common rules are required). As a result, effective freezing and confiscation of assets would not be possible in all cross-border cases.

9. Implementation, monitoring and Evaluation

The implementation of the preferred option should be subject to future monitoring and evaluation.

This report has repeatedly highlighted the lack of statistical data on asset confiscation and the poor quality of available data. As a result of these data gaps, it is currently not possible to carry out a proper evidence-based assessment of the impact of new policies/legislation at EU level or at Member State level in most countries. In addition, information on the extent to which mutual recognition facilitates cross-border enforcement is not readibly available as a result of which the role of mutual recognition is poorly understood.

For this reason, the preferred option includes the introduction of reporting obligations on the Member States in relation to asset confiscation work. Data will be collected by judicial authorities (courts, prosecution offices) asset management offices and other authorites in charge of asset disposal, at least on an annual basis[35]. The data so collected will feed into monitoring and evaluation activities and will allow the Commission to assess to what extent the proposed legislation achieves its objectives. Particular attention should be paid to those Member States where data collection is relatively under-developed. Examples of the type of data that could be collected include:

· Number of freezing orders executed

· Number of confiscation orders executed

· Value of assets frozen

· Value of assets recovered

· Number of requests for freezing orders to be executed in another Member State

· Number of requests for confiscation orders to be executed in another Member State

· Value of assets recovered following execution in another Member State

· Value of assets destined to social use

· Number of cases where confiscation is ordered/ number of convictions for the criminal offences covered by the Directive. 

In order to monitor the effective implementation of the proposed legislation the Commission will prepare an implementation plan and produce regular implementation reports based on consultations of the Member States and stakeholders. The first report is in principle foreseen three years after the entry into force of the legislation. The mapping exercise of the asset confiscation legislation in the Member States which was carried out in preparation for the present impact assessment could be used as a baseline for monitoring developments in law and utilisation in the Member States.

Evaluations will also be carried out on a regular basis, the first report being foreseen five years after the entry into force of the legislation. The evaluation reports could include a cost-benefit modelling exercise to assess the current and estimate the future profitability of asset confiscation work.

Transposition workshops and other expert meetings will also take place to discuss implementation problems. The exchange of best practices in all the phases of the confiscation process will continue to take place within the EU Asset Recovery Offices Platform.

  Annex 1                     Summary of fieldwork of the external IA study

Fieldwork was carried out in situ and/or by telephone in all 27 EU Member States.  In order to understand how Member State laws operate in practice through the EU, government practitioners (i.e. police, prosecutors and others) were interviewed in each Member State. Only in Poland did scheduling difficulties prevent this. This core of interviews was complemented with other perspectives from judges, defence lawyers, academics and, in the Italian case, from persons with experience in the social reuse of confiscated assets.  Table 1 summarises this fieldwork.

Table 1 Fieldwork in Member States

Moreover interviews were held with representatives of EU and international institutions:

· Europol

· Eurojust

· CARIN (the Camden Asset Recovery Inter-Agency Network)

· Moneyval

· FATF

· Council of Europe: European Court of Human Rights

· Council of Europe: Venice Commission

Annex 2                      Confiscation statistics

Statistics are presented for the following Member States where relevant material was provided or located through fieldwork and data search:

· Bulgaria

· France

· Germany

· Hungary

· Ireland

· Italy

· Netherlands

· United Kingdom

Bulgaria

Table 2 Bulgarian Statistics, 2006-2010

|| Freezing cases p/a || Assets frozen p/a (€m) || Confiscation cases p/a || Costs* (€ millions)

2006 || 100 || 21.8 || 12 || .

2007 || 109 || 66.6 || 33 || .

2008 || 126 || 66.3 || 57 || .

2009 || 155 || 254 || 79 || 6.5

Source: CEPACA Annual Report (2009).

As at the end of 2009, of all the confiscation cases brought to date:

· 133 remained at first instance trial,

· 28 decisions at first instance (CEPACA won 22),

· 13 decisions at second instance (CEPACA won 7),

· 6 cases finalised (CEPACA won 4).

Recovered assets from the 4 cases won = €1.0m.

Value of assets in the 29 successful cases = €10m.

France

Table 3 French statistics (2005-2009)

|| Freezings by police (€m) || Freezings by Gendarmerie (€m) || Total freezings (€m)

2005 || . || . || 51.3

2006 || 60.5 || 11.4 || 71.9

2007 || 51.8 || 3.8 || 55.5

2008 || 35.1 || 58.8 || 93.9

2009 || 58 || 127.7 || 185.7

Source: Reports of PIAC (Platform for Identification of Criminal Assets)

Germany

Table 4 German statistics, organised crime (1992-2009)

|| % of O.C. investigations in which assets seized || Estimated O.C. profit in these cases (€m) || Amount seized (€m) || Total number of  recorded crimes

1992 || 5.0% || . || . || .

1993 || 6.6% || . || || .

1994 || 6.8% || . || || .

1995 || 8.3% || . || <20 in mid 90s || .

1996 || 10.5% || . || || 6,647,598

1997 || 12.1% || . || || 6,586,165

1998 || 21.5% || . || . || 6,456,996

1999 || 22.2% || . || 118.5 || 6,302,316

2000 || 30.2% || . || . || 6,264,723

2001 || 30.7% || 760 || . || 6,363,865

2002 || 25.0% || 1,500 || 31 || 6,507,394

2003 || 25.3% || 468 || 69 || 6,572,135

2004 || 24.2% || 1,337 || 68 || 6,633,156

2005 || 25.4% || 842 || 97 || 6,391,715

2006 || 25.9% || 1,815 || 60 || 6,304,223

2007 || 29.1% || 481 || 39 || 6,284,661

2008 || 27.0% || 663 || 170 || .

2009 || 26.9% || 903 || 113 || .

Source: Utilisation, freezing, profit (BKA annual organised crime situation reports); Total number of crimes (Eurostat, 2010)

It is important to note that utilisation, amount seized and estimated profit refer to organised crime as defined in the BKA. We do not have a precise definition of ‘profit’ in this context.

Table 5 German statistics, all crime (1999-2009)

|| Number of proceedings in which assets confiscated || Total (state and civil) claim (€m) || Total amounts confiscated or forfeited (€m) || Total number of  recorded crimes

1999 || - || - || 219 || 6,302,316

2000 || - || - || .536.9 || 6,264,723

2001 || - || - || .332.6 || 6,363,865

2002 || - || - || 294 || 6,507,394

2003 || - || - || - || 6,572,135

2004 || 6045 || 1268 || 306 || 6,633,156

2005 || 6010 || 1191 || 319 || 6,391,715

2006 || 6101 || 1066 || 301 || 6,304,223

2007 || 7050 || 592 || 219 || 6,284,661

2008 || - || - || - || -

2009 || 6725 || 901 || 281 || -

Source: Confiscation statistics (Fieldwork, FATF 2010, Fijnaut & Paoli (2004) Organised Crime in Europe pp752-753); Total number of crimes (Eurostat, 2010)

Hungary

Table 6 Hungarian statistics (1999-2008)

|| Recorded crimes || Convictions (total) || Convictions (property and financial crime) || Forfeiture cases || Amount frozen/seized (€m)

1999 || 505,716 || 95398 || 50840 || 56 || .

2000 || 450,673 || || || || .

2001 || 465,694 || 94538 || 48249 || 14 || .

2002 || 420,782 || || || || .

2003 || 413,343 || 93442 || 45090 || 35 || .

2004 || 418,833 || || || || 41

2005 || 436,522 || 97558 || 44676 || 233 || 69

2006 || 425,941 || || || || 42

2007 || 426,914 || 86705 || 38112 || 598 || 102

2008 || . || || || || 57

Source: Utilisation, Amount frozen seized (Police interviews and Criminality and Criminal Justice' report of Hungarian Prosecutor General, 2008); Recorded crimes (Eurostat, 2010)

These statistics evidence a rising utilisation rate (forfeiture cases as a function of convictions). Data is not available for amounts ordered confiscated or subsequently recovered.

Ireland

Table 7 Irish statistics (2003-2009)

|| CAB recovery from NCB confiscation (€m) || CAB recovery from revenue powers (€m) || CAB total recoveries (€m) || Running costs of CAB (€m) || Recorded crimes

2003* || ? || 10 || ? || 5.7 || 103,462

2004* || ? || 16.4 || ? || 5.7 || 99,244

2005 || 2 || 16.3 || 18.3 || 5.2 || 102,206

2006 || 3 || 19.1 || 22.1 || 5.2 || 103,178

2007 || 0.3 || 10 || 10.3 || 5,1 || .

2008 || 6.1 || 5.9 || 12.0 || 7.5 || .

2009 || 1.4 || 5.2 || 6.6 || 6.9 || .

 Source: CAB data (Annual Reports of the CAB); Recorded crimes (Eurostat, 2010).

The recovery data relates only to the Criminal Assets Bureau (CAB) which has non-conviction based confiscation powers and also revenue powers (i.e. the ability to levy tax on previously undeclared income where even a non-conviction based case cannot be made out on the evidence.  Amounts recovered from non-conviction based confiscation mostly relate to work from previous years due to a lag between the seizure of assets and their vesting in the state (unless there is disposal by consent, the law requires seven years).[36] Monies recovered by victims due to the work of CAB were not identified and so were not available to add to these figures.

Operating costs for the CAB include the cost of training regular Gardi (police officers) so that conviction-based confiscation can be performed at local level.  No conviction based data is available.

Italy

Table 8 Italian statistics (1992-2009)

|| Assets investigated || Assets ordered  confiscated || Final orders || Disposals || Recovered value (€) || Social reuse (€m)

1992 || || 0 || 13 || 9 || 1.8 || 0.5

1993 || || 85 || 9 || 3 || 0.4 || 0.1

1994 || || 1 || 27 || 2 || 0.2 || 0.1

1995 || || 0 || 22 || 5 || 1.0 || 0.5

1996 || || 15 || 102 || 18 || 3.7 || 2.7

1997 || || 71 || 340 || 63 || 18.6 || 9.3

1998 || || 155 || 404 || 129 || 18.2 || 8.6

1999 || || 392 || 640 || 216 || 37.2 || 27.2

2000 || || 435 || 575 || 249 || 38.9 || 18.2

2001 || || 203 || 718 || 231 || 47.4 || 35.4

2002 || || 211 || 477 || 329 || 89.9 || 71.5

2003 || || 464 || 300 || 287 || 40.8 || 22.5

2004 || || 660 || 328 || 287 || 47.4 || 27.1

2005 || || 1044 || 400 || 190 || 51.8 || 32.1

2006 || 4427 || 1566 || 414 || 172 || 31.8 || 10.0

2007 || 8040 || 1790 || 325 || 518 || 97.5 || 38.1

2008 || 6173 || 949 || 319 || 804 || 165.5 || 80.8

2009 || 12741 || 2333 || 380 || 544 || 101.3 || 60.7

TOTAL || 62551 || 11067 || 6207 || 4074 || 797.1 || 447.4

 Source: Italian Department of Justice (2010)

Social reuse data refers to assets used or allocated for social purposes by municipalities.  It does not include and assets or revenue streams allocated to law enforcement agencies. 

Netherlands

Table 9 Dutch statistics (2003-2009)

|| Frozen assets under administration (€m) || Amount ordered confiscated (€m) || Amount recovered (€m) || Recorded crimes

2003 || . || . || 10 || 1,369,271

2004 || . || . || . || 1,319,482

2005 || . || . || . || 1,255,079

2006 || . || . || . || 1,218,447

2007 || . || . || . || 1,214,503

2008 || 550 || . || 23.4 || .

2009 || 600 || 70 || 50 || .

Source: Amount confiscated (Authors’ fieldwork) ; Recorded crimes (Eurostat, 2010)

Data for frozen assets under administration includes assets frozen in previous years and remaining under administration. Data for amounts received refers to confiscation orders successfully enforced, which typically relate to confiscation orders from previous years.

United Kingdom

Table 10 UK statistics (2001-2009)

|| Amount confiscated (£m, realised orders) || Recorded crimes

2001 || . || 6,085,903

2002 || . || 6,544,490

2003 || 25 || 6,548,691

2004 || 46 || 6,193,756

2005 || 84 || 6,096,153

2006 || 125 || 5,968,674

2007 || 136 || 5,444,648

2008 || 146 || .

2009 || 154 || .

Source: Amount confiscated (UK Home Office); Recorded crimes (Eurostat, 2010)

Annex 3                      List of EU-level actions

In order to meet the specific and operational objectives identified and remedy the shortcomings resulting from the problem definition, this impact assessment proposes an analysis of 21 EU-level policy actions (some of which are complementary) targeting particular operational objectives. They are described below, grouped by the specific objectives to which they relate. 

Further harmonising the confiscation powers

1.   Promoting implementation of existing confiscation obligations.  Although the trends towards compliance with FD 2005/212/JHA are positive, the European Commission could help to ensure ongoing progress via continued implementation/expert workshops.

2.   Confiscation of all valuable benefits, including indirect proceeds. The EU legal framework could harmonise a definition of criminal ‘proceeds’, to ensure the recovery of ‘indirect’ proceeds resulting from the appreciation in value, or profitable reinvestment, of direct proceeds.  Harmonisation could also ensure that any valuable benefit (including, for example, the value of liabilities avoided) is liable to confiscation.

3.   Civil standard of proof regarding whether an asset is "criminal".  The standard of proof on whether particular assets are proceeds could be harmonised to a lower "balance of probabilities" standard, to make it more difficult for convicted criminals to retain type 1 assets.

4.   Separate confiscation proceedings.  The EU legal framework could ensure that separate confiscation proceedings can be brought also at a later date when criminal proceedings are finalised.

5.         Strengthening extended confiscation. The EU legal framework could be simplified and strengthened by providing for extended confiscation at least where a court finds it substantially more probable that the assets of a person convicted of an offence covered by Article 83(1) TFEU are derived from other similar criminal activities. 

6.   Non-conviction based confiscation in limited circumstances.  The EU legal framework could make ordinary confiscation possible in circumstances where a conviction cannot be obtained because the suspect has died, fled the jurisdiction, or is unable to stand trial.

7.   Third party confiscation.  Laws could be harmonised by requiring third party confiscation to be available for assets received for less than market value and which a reasonable person in the position of the third party would suspect to be derived from crime. 

Harmonising the preservation powers

8.   Universal freezing. Harmonised minimum standards for freezing could ensure that it is possible to preserve any assets and would ease the mutual recognition of freezing orders.

9.   Mechanisms to safeguard freezing.  Member States could be required to have in place appropriate mechanisms to ensure that assets in danger of being hidden or transferred out of the jurisdiction are able to be immediately frozen/seized.  This would include, in appropriate circumstances, the ability to freeze/seize prior to seeking a court order.

10. Powers to realise frozen assets.  Harmonisation could ensure that, regardless of how frozen assets are managed, there are powers to realise them at least where they are liable to decline in value or uneconomical to maintain.

11. Designating Asset Management Offices (AMOs). Further harmonisation could require all Member States to entrust the management of frozen assets to AMOs at a national or regional level. This could increase efficiency and promote best practice.

Enhancing the enforcement powers (mutual recognition)

12. Promoting implementation of existing mutual recognition obligations.  The Commission could help to ensure ongoing progress in implementing Framework Decisions 2003/577/JHA (freezing orders) and 2006/783/JHA (confiscation orders) via implementation/expert workshops. 

13. Mutual recognition of all types of orders.  The EU legal framework could remove existing limitations on the mutual recognition of freezing and confiscation orders, allowing orders to better circulate around the Union.  This would also make the legal framework more coherent.

14. Mutual recognition of compensation orders.  The legal framework could be simplified by consolidating FD 2006/783/JHA and 2005/214/JHA and extending their scope to include all compensation orders made in the context of criminal proceedings.

Enhancing the utilisation of existing tools

The existing EU legal framework neither obliges utilisation nor provides for incentives for cultural change through the normalisation of asset confiscation activity (where failure by police and prosecutors through negligence or reticence to recover criminal assets would be perceived as unacceptable).

15. Utilisation workshops. Utilisation workshops on the profitability of asset confiscation work to the benefit of government decision-makers in some Member States could increase utilisation of these tools and provide a forum for the sharing of scientific knowledge and practitioner experience.

16. Reporting obligations. Reporting obligations could be introduced, for example an obligation to report, for all crimes covered by TFEU article 83(1), assets frozen, the confiscation orders (if any) obtained and the type of order. This would also help generating statistics which could be used for evaluation purposes.

17. Mandatory assets investigation. The EU legal framework could require investigators to open a parallel financial investigation, at least for the crimes listed in TFEU article 83(1). 

18. Limited judicial discretion. Judicial discretion could be limited by requiring freezing to be ordered wherever there is reasonable cause to suspect that an asset may become liable to confiscation and, in the event of a criminal conviction, by requiring confiscation to be ordered unless doing so would disproportionately affect fundamental rights.

19. Consolidated mutual recognition forms. A single form for all types of mutual recognition at the investigative stage could be provided within the European Investigation Order (and by suppressing the existing mutual legal assistance alternative).

20. Enforcing the primacy of mutual recognition. The EU legal framework could suppress the use of mutual legal assistance with respect to freezing and confiscation by repealing the existing mutual legal assistance conventions as regards requests between Member States .

Annex 4          Asset recovery in the UK

Utilisation trends

As with other EU Member States, the UK’s traditional approach to criminal justice has been to detect and prosecute offenders, punishing them with fines and imprisonment.  This approach came under scrutiny following the Cuthbertson case, in which a drug trafficker sentenced to a lengthy jail term retained £750,000 in proceeds because the prevailing forfeiture regime in the Misuse of Drugs Act 1971 was too narrow.  This led eventually to stronger asset recovery laws in the form of the Drug Trafficking Offences Act 1986 and the Criminal Justice Act 1988.  These laws extended confiscation to all indictable offences and introduced value confiscation.  Extended confiscation—a reverse burden of proof regarding the legitimacy of all assets acquired in the preceding 6 years—was introduced for drug offences and then generalised by the Proceeds of Crime Act 1995.[37]

Visible impacts can be expected to lag the introduction of such new powers for several reasons: it takes time for practitioners to learn how to use them; it takes time for cases to progress through the courts; legal challenges will further slow the first wave of cases.  Yet Levi and Osofsky reported in 1995 that confiscation powers were still being utilised only occasionally for drug crimes, and rarely for other crimes (Levi and Osofsky, 1995).  Five years later, the Performance and Innovation Unit of the Cabinet Office reported that:

In the last five years, confiscation orders have been raised in an average of only 20 per cent of drugs cases in which they were available, and in a mere 0.3 per cent of other crime cases. The collection rate is running at an average of 40 per cent or less of the amounts ordered by the courts to be seized. Specially tasked law enforcement officers struggle to investigate the financial aspects of crime to support this effort, but their effectiveness is limited by their numbers and modest training.[38]

This report’s recommendations included: a strategic approach aimed at incentivising asset recovery work within practitioner communities, more resources for financial investigation and a ‘new legislative attack’.  The latter took the form of the Proceeds of Crime Act 2002, which consolidated existing legislation, tightened some aspects and introduced three new elements: a non-conviction based ‘civil recovery’ power, a non-conviction based cash seizure/forfeiture regime and new revenue powers to allow otherwise unrecoverable criminal profits to be taxed.[39]  These new powers were given over to a new Asset Recovery Agency (ARA), while responsibility for conviction-based confiscation work remained with police forces throughout the country.  A year after the new legislation entered into force, a government report concluded that there were:

pockets of excellent practice but that the overall application of the powers across England and Wales was patchy, with money laundering and confiscation seen as complex, specialist activities, divorced from mainstream business. Activity was often only targeted at the higher profile ‘crime barons’ and almost exclusively against drug trafficking, leading to failure to use POCA to its full potential. Opportunities to combat those engaged in volume crime, street robbery and low-level drug dealing were being missed.[40]

Essentially, whereas ARA had embraced asset recovery as its raison d’être, it remained alien to the mindset of ordinary police officers and prosecutors.  Part of the solution, beginning in April 2004, was to raise utilisation within all relevant government agencies through the Asset Recovery Incentivisation Scheme (ARIS), whereby 50% of the revenue stream generated by confiscated assets is returned to the agencies who played a role.  Another part of the solution has been a concerted effort to train and deploy financial investigators.  These efforts have led to increased utilisation, with more than 10% of Crown Court convictions for acquisitive crimes (fraud, burglary, drug trafficking, etc) now resulting in confiscation orders.

Figure 11: Utilisation in the UK Crown Court, April 2006 – March 2010

Source: Data from JARD and other sources, collated by the National Policing Improvement Agency

Although a utilisation rate just above 10% may seem low, the effective utilisation rate will be somewhat higher because these total figures included cases in which a confiscation order would be inappropriate, either because there are no relevant proceeds (despite the offence being of an acquisitive type) or because there are known to be no recoverable proceeds (e.g. where the proceeds have been dissipated).  Against this, there are also cases in which ‘nominal’ confiscation orders in the amount of £1 are obtained, to allow the issue to be reopened should proceeds be identified at a later stage.

Another interesting point to note is that, in the last four years, recoveries in the Crown Court have been rising in absolute terms (by an average of more than 12% a year) but not as a proportion of convictions for acquisitive crimes in the Crown Court, because these too are rising.[41] However, even though the 2006–2010 Crown Court time-series data does not show an increasing rate of utilisation, a look at past statistics (for example the 0.3% utilisation rate for non-drug cases quoted above) suggests an increase in the wake of POCA 2002.  Indeed, there is strong evidence of this in the form of hugely increased treasury receipts, which reached £154m in FY2009/10 (see Figure 12). This figure is, however, felt by the UK government to be still too low. In particular, a recent report has bemoaned the UK law enforcement community’s failure to ‘mainstream’ asset recovery work. Significantly, it recognised that one of the main barriers may be profitability:

"There is a dichotomy between the need to mainstream asset recovery if the value recovered from confiscation is to grow significantly, and the risk that a move away from specialisation could dilute skills, knowledge and experience, and prejudice performance if it is not done in a carefully planned manner. One route out of the conflict would involve a significant commitment to training and performance management over a sustained period, in order to achieve the necessary shift in thinking amongst frontline staff in all agencies. Alternatively, the way forward is to recognise that mainstreaming is unlikely to provide value for money, and focus resources where they will be most cost-effective, such as in expanded specialist units. There is also an argument for making the statutory process leading to a confiscation order more streamlined, so that orders take less time, and there are fewer procedural steps to take; this could improve the cost-effectiveness and the commitment to asset recovery at the same time."[42]

We now turn to consider the profitability of asset recovery work in the UK.  We take a narrow approach, looking only at the ongoing costs of asset recovery work and the annual revenue stream generated, disregarding the value of any other potential economic, social and environmental benefits.  We focus on ongoing costs.  We lack the data to examine one-off costs, which we estimate to be small by comparison, especially given the period under scrutiny (i.e. several years after the introduction of the POCA 2002).

9.1.1. Profitability analysis

The UK maintains a Joint Asset Recovery Database (JARD) which records all amounts finally recovered in favour of the state (though not those recovered in favour of victims).  Records date back to the financial year April 2003–March 2004 following the introduction of POCA 2002.  Amounts are net of any management expenses payable to private receivers, but not of agency operating costs.  The data in Figure 12 shows a clear upward trend, reflecting increasing utilisation of powers in recent years, reaching £154m in FY2009/10.

Figure 12: Assets recovered in favour of the state, England and Wales, 2001-2009

Source: UK Home Office 2003-2009, interview estimate for 2001

In order to analyse profitability we now turn to consider the ongoing costs of asset recovery work within the UK.  In the absence of an equivalent system to JARD for recording costs, we examine the constituent parts of the UK’s asset confiscation apparatus.  In many cases, the agencies concerned have a reasonably good understanding of their own costs, as there has been considerable emphasis placed on this politically.[43]  Indeed, understanding costs is essential when negotiating the division of assets returned as incentives under ARIS.[44]  In some cases, agencies have published information which directly addresses costs and profitability.  In other cases, we base our estimates on expert opinions elicited through fieldwork with senior members of the agencies themselves.  Using this information, we are able to roughly estimate the ongoing cost of the UK’s asset confiscation apparatus.

We begin by reviewing the main ‘frontline’ agencies involved in asset recovery work, examining available cost data and making assumptions where necessary along the way.

· Police authorities.  Police authorities are responsible for financial investigations in support of criminal confiscation proceedings pursuant to POCA section 6, and also (using their own legal representation) for the cash seizure/forfeiture procedure in POCA Part 5.  The UK has some 50 police authorities, all of which are more or less engaged in asset confiscation work, using financial investigators who receive the standard training.  The Metropolitan Police is by far the largest force.  Data for the 2009/10 financial year shows that it spent £10.7m on asset confiscation activity—including £500,000 funding for community programmes—whilst it had receipts of £10.9m generated by asset confiscation activity itself (calculated under ARIS as 50% of forfeited cash plus 18.75% of conviction-based recoveries).[45]  We assume similar levels of profitability for other police authorities.

· Her Majesty’s Revenue and Customs (HMRC).  HMRC has an equivalent role to the police authorities, for cases within its area of responsibility.  In the absence of publicly available data, we make the same assumption for profitability as per the police authorities, i.e. that it is equivalent to that of the Metropolitan Police.[46]

· The Serious and Organised Crime Agency (SOCA).  SOCA’s asset recovery work includes conviction-based proceedings arising from its own investigations into serious and organised crime (an equivalent role to the police authorities), non-conviction based ‘civil recovery’ cases, exercise of revenue (taxation assessment) powers and, where it seizes cash in the course of an investigation, the non-conviction based cash seizure/forfeiture process. The civil recovery and revenue work was previously undertaken by the Asset Recovery Agency (ARA), which was merged into SOCA from 1 April 2008. As a sui generis entity administering a complex piece of legislation, the ARA was beset with lengthy judicial processes and never became ‘profitable’ in the sense that its costs exceeded the income stream from its asset recovery work in all five years of its existence. Recently, SOCA’s 2008/09 accounts have been audited in a way which specifically permits comparison with the work previously performed by ARA (civil recovery, taxation and some ‘legacy’ criminal confiscation cases).[47] In these comparable areas (which represent the majority of SOCA’s asset recovery work), SOCA recovered £20.2m at a cost (including receivers’ fees) of £16.3m. Having not obtained any data regarding the profitability of the balance of SOCA’s asset recovery work (additional conviction-based cases), we assume an equivalent level of profitability.

· The Crown Prosecution Service (CPS). CPS brings conviction-based confiscation proceedings on the back of investigations by the police, HMRC and SOCA, and also works to enforce some of the more complex orders obtained.[48]  On 1 January 2010, the CPS absorbed the Revenue and Customs Prosecution Office (RCPO), which previously brought confiscation cases on behalf of HMRC.  Expenditure on asset recovery work is not published; our own fieldwork (conversations with experts) suggests that it is approximately equal to CPS’s share of ARIS revenue.  An unpublished internal audit of RCPO undertaken prior to its merger with CPS suggested that its activities had previously been somewhat less profitable, though no figures are given.[49]

· Her Majesty’s Courts Service (HMCS). HMCS enforces the majority of conviction-based confiscation orders.  This work involved writing letters, fixing hearing dates, and then taking measures following the activation of default judgment by a Magistrate.  An unpublished study showed that in 2008/09, HMCS spent slightly less on asset recovery work than the ARIS funding it received (at the rate of 12.5% of the value of the orders enforced). [50]

The roles of the agencies described above (which are the main agencies administering POCA) are summarised in Figure 13. The assessment of profitability can be summarised as follows:

· Police, HMRC, CPS and HMCS: approximately funded with ARIS funding.

· SOCA: recovers a little more than what it spends, but not enough to be funded through its share of ARIS.

Broadly speaking, there are two plausible explanations for SOCA work being less profitable than that of the other frontline agencies.  First, mentioned already, is that SOCA administers a sui generis regime which generates an additional legal burden as case law must be generated, at significant expense in terms of legal fees.  Second¸ SOCA generates less income through cash seizure/forfeiture powers than the other investigative agencies.  These powers are more profitable than other powers because they involve abbreviated court procedures, with the entire 50% of ARIS funding going to the investigative agency.  A third explanation—higher overheads due to smaller agency size, is less relevant following the ARA’s merger into SOCA.

Figure 13: Functions of ‘frontline’ UK asset confiscation agencies

|| Investigation || Confiscation || Enforcement

Criminal confiscation (including extended)— POCA s6 || Police; HMRC; SOCA || CPS || HMCS; CPS

Civil recovery— POCA Part 5 Chapter 2 || SOCA

Cash seizure/forfeiture— POCA Part 5 Chapter 3 || Police; HMRC; SOCA

Taxation— POCA Part 5 Chapter 6 || SOCA

 

Because the foregoing estimates are expressed as fractions of amounts recovered and ARIS receipts, an absolute cost estimate requires disaggregated recovery data.  Available data for the 2008/09 financial year is provided in Table 2, for financial year 2008/09.

Figure 14 Disaggregated treasury receipts, FY 2008/09

Agency || Cash Forfeiture || Confiscation [with CPS/HMCS] || Civil Recovery & Taxation || Total

Police || £27.51m || £54.03m || - || £81.54m

HMRC || £10.51m || £18.91m || - || £29.42m

SOCA || £1.78m || £10.05m || £16.83m || £28.66m

Other || - || £6.09m || £2.29m || £8.38m

Total || £39.80m || £89.08m || £19.12m || £148m

It can be seen that the non-SOCA share of confiscation work amounts to some £120m, including £8m ‘other’, which we will assume to be similarly profitable to work undertaken by the non-SOCA agencies.[51]  Based on the foregoing assumptions, we therefore calculate the ongoing annual cost of asset confiscation work performed by frontline agencies in the UK (specifically, in England and Wales) in 2008/09 to be:

119.34 * 0.5 + 28.66 * (16.3 / 20.2) = £82.8m

In light of the numerous assumptions which have been made (in particular around the police authorities and HMCS), it is appropriate to express this amount as a range with ±15% uncertainty, i.e. between £70.4m and £95.2m.

To obtain a complete picture of the costs of the administration it is also necessary to consider other costs not borne by frontline agencies.  The main such cost is that of an increased caseload for the court system.[52]  This cost is not accounted for within the foregoing analysis, where the profitability analysis for HMCS refers only to enforcement work, and not the cost of hearing cases.

We begin by considering criminal confiscation cases, which are typically heard in the Crown Court.  In the absence of any more specific data, we assume that the cost of such a case is the same as the cost of the average Crown Court case.[53]  In 2009, the Crown Court dealt with 147,200 cases, ranging from guilty pleas to lengthy jury trials.[54]  In 2007/08, the cost of operating the Crown Court was calculated to be £382m.[55]  This amounts to some £2,600 per case.  Statistics for 2008/09 indicate that there were 4717 confiscation orders made in the Crown Court that year; this amounts to a total cost of:

2,600 * 4,717 = £12.2m

In addition, it is necessary to account for the costs of civil confiscation cases and taxation cases (both brought by SOCA in civil courts), as well as for cash seizure/forfeiture cases.  The former may be less costly to the court system than conviction-based cases because civil courts charge fees, with the aim of making civil procedure cost-neutral to the state.  The latter may also be less costly, because an abbreviated procedure is employed vis-à-vis conviction-based confiscation, and the cases are able to be heard in the Magistrates Court, which has lower operating costs.  On the other hand, these non-conviction based cases have raised many new questions of law which have been appealed to higher courts, causing much additional delay and expense.  Overall, therefore, we make the assumption that these cases present the court system with a similar level of cost to conviction-based cases.  Applying a ratio of 148:89.1 (based on table 2) we therefore calculate the overall ongoing annual cost upon courts as:

12.2 * (148 / 89.1) = £20.3m

Again, in light of the broad-ranging assumptions which have been made, it is appropriate to express this amount as a range with ±15% uncertainty, i.e. between £17.3 and £23.4m.

Summing front-line agency and court costs, we arrive at the following estimate of overall annual ongoing cost (and thus profitability, based on a known return of £148m) of asset recovery activity in England and Wales for 2008/09.  These calculations are set out in Figure 15.

Figure 15 Cost and profit calculations for FY 2008/09

Element || Low estimate || High estimate

Cost (m£) || Profit (m£) || Cost (m£) || Profit (m£)

Front-line agencies || 70.4 || || 95.2 ||

Courts || 17.3 || || 23.4 ||

Total || 87.7 || 60.3 || 118.6 || 29.4

To be sure, this analysis has examined the profitability of asset recovery activity on FY 2008/09 only: being the sixth year after the introduction of POCA 2002, and the fifth year after the introduction of ARIS.  Receipts to the state in that year flow from a ‘pipeline’ of work which includes many cases commenced in previous years; similarly, many cases commenced in that year will not emerge from the pipeline until future years.[56]  The time taken for cases to progress through the pipeline varies greatly: cash seizure/forfeiture cases in the Magistrates Court typically take 3 to 6 months (and are often only lightly contested), whilst conviction-based cases in the Crown Court can take several years (until appeal rights are exhausted), with convicted criminals often fighting hard to retain their wealth.  An important corollary is that, whereas asset recovery work in the UK appears now to be profitable, it was not necessarily immediately profitable in the wake of POCA 2002, due to: the lag in building a pipeline of work from an initially low base; the costs associated with establishing the ARA from scratch; and the costs of answering the legal challenges which inevitably followed the introduction of novel powers.  Unfortunately, whilst the ARA’s financial statements are public, a lack of data for other elements of the system prevents us from estimating profitability in these early years.

Future possibilities

We have seen that POCA 2002 and the attendant focus on utilisation has led to increased asset recovery work, the current level of which is profitable in the (narrow) sense that receipts into government coffers exceed the total cost of the work itself.  For the purpose of impact analysis, it is useful to consider now the potential for continued growth.  This requires estimates of future recoveries and costs.

Turning first to future recoveries, we begin by examining the available time-series data in order to estimate the relationship between past amounts collected and current amounts over the period 2003 to 2009.  Statistical tests suggest the amount recovered in the current year is correlated with the previous year’s amount recovered.[57]  As such, we regress the amount recovered in the current year on the amount recovered in the previous year (and a constant). Using the mean point estimate for the relationship, we compare the actual and predicted (or estimated) amounts of asset recovery in each year from 2003 to 2009.  As shown in Figure 16, we can see that the match between actual and estimated amounts coincide better in more recent years. There are any number of explanations for this, not least of which is the short time-series.

Figure 16: Illustration of model estimates and actual values

Source: Authors.

Estimates suggest that we can be 95 percent confident that the mean proportion of last year’s recovery associated with this year’s is greater than 50 percent and less than 110 percent. Given that there are many uncertainties and other factors for which we have not accounted, we use this range (rather than the point estimate we used to compare our model to the actual amount recovered) to determine the possible range of amounts collected in the UK through to 2020.  As shown in Figure 17, we find that there may be between £80 million to £1.2 billion collected in ten years time.

Figure 17: Potential recoveries through to 2020, United Kingdom

Because these estimations are based purely on the time-series model, it is useful to discuss the type of scenarios which they represent.  This is particularly so for the maximal prediction, which involves year-on-year increases which do not diminish with time, thus assuming not only that the pool of available criminal assets is large enough to support this, but also that the marginal return on additional investment remains constant.  The first assumption seems likely to be true, given that the annual turnover of organised crime in the UK is estimated to be £15b, and given also that extended confiscation powers render previous years gains liable to confiscation.  The second assumption demands closer examination.

All of the frontline agencies involved in financial investigation and bringing confiscation proceedings have finite resources, necessitating selectivity.  Managers and practitioners must decide how much confiscation work to undertake and also which cases to prioritise.  These decisions should follow a harm-reduction ethos, which should involve (at least for police) differing approaches in different localities with different problems. Sometimes, authorities may take on unprofitable cases in order to deal with specific problems—an example from the UK (and other Member States) is the confiscation of expensive cars from low-ranking criminals in order to discourage crime within their communities.  Generally, however, the very purpose of asset confiscation justifies focusing upon profitable cases (especially those where the assets are more readily recoverable), as the deterrent effect of confiscation orders is largely a function of the amount recovered.[58]  A more cynical view is that ARIS may encourage agencies to focus on high-value cases at the expense of a harm-reduction ethos.  In any event, it is reasonable to assume a bias towards ‘low hanging fruit’—i.e. that the most profitable cases tend to be selected ahead of intractable or low-value cases.[59]  All else being equal, this will cause the marginal (and overall) profitability of asset recovery work to decline as more work is undertaken.  Some countervailing  trends will, however, tend to negate this effect.  Frontline agents will become more efficient at identifying, freezing, confiscating and recovering assets due to learning effects and economies of scale.  Court processes will become more efficient for similar reasons, and also because legal challenges will be fewer as the law becomes more settled.  Against these  trends, success will be met with increased efforts to hide wealth (as criminals play the ‘game’ against the state, making asset recovery work more expensive.

Ultimately there must come a point at which the profitability of asset recovery work begins to decline (because all the low-handing fruit has been picked), and another at which further efforts are unprofitable.  In the absence of a detailed model, however, we have no better guide than expert practitioner opinions.  In this regard, whilst it is recognised that not every case of acquisitive crime will present an opportunity for profitable asset recovery (hence the doubt, about ‘mainstreaming’ asset recovery work expressed in the recent Joint Thematic Review, and discussed above), there seems to be a general concensus that much more could be profitably done.  Some experts favour a more systematic use of money laundering laws and confiscation laws to target top-tier criminals.  Others consider that much could be achieved if police simply did ‘more of the same’ by employing more financial investigators in more of the existing investigations into known mid-ranking criminals.  It is also believed that financial investigation exposes new crimes and criminals, increasing the pool of assets practically available to be targeted.  These opinions have one thing in common: that a lack of trained financial investigators is a limiting factor, and will remain so into the foreseeable future.[60]

Turning now to potential future costs, these are plotted in Figure 18 as minimum and maximum scenarios. The data for 2008 are the low and high estimates in Figure 17, with previous years’ data (in the absence of any estimate) assumed to be in the same ratio, but reduced in proportion to the relative amount recovered in each year versus 2008. Looking forward, the minimum scenario then assumes that costs remain at 2009 levels, i.e. that asset recovery operations do not expand. The maximum scenario assumes instead that costs escalate at 15% per year as more and more financial investigators and other agents are hired, and more cases brought.[61]

Figure 18: Potential cost of asset recovery through to 2020, UK

Finally, we combine minimum and maximum revenue and cost forecasts to produce minimum and maximum profit forecasts, shown in Figure 19. It should be remembered when interpreting these data that the maximum scenario assumes that there are no limitations upon the profitability of asset recovery work through to 2020, whereas the minimum scenario assumes fixed costs and profitability which (pursuant to the formula derived from the historical time-series) declines and then plateaus.

Figure 19: Potential profitability to 2020, UK

Source: Authors’ calculation.

Annex 5          EU27 profitability model

This Annex includes profitability estimates for the EU27 by using proxy indicators to generalise the UK estimate in Annex 6. This approach is made necessary by the paucity of useful empirical data, especially as regards the cost of asset confiscation work.  This involves:

· devising a logic model by adverting to available evidence about the causes of (non)utilisation;

· identifying proxy indicators and available EU27 datasets for the identified barriers and drivers within the model;

· using the proxy indicators to generate output from the model; and

· interpreting the output data.

It is worth reiterating the underlying assumption that the UK situation in 2008/09 approximates the potential impact by 2020 of the maximal policy option (without mutual recognition) upon a Member State with a currently low rate of utilisation.  We shall return to the implications of this, and other assumptions, when interpreting the results.  [?]

The starting assumption is the basic equation that profit = income – cost.  Lacking any EU-27 indicators of the cost of asset confiscation work, we assume this to be a function of overall administrative efficiency in each Member State.  Likewise, lacking any EU-27 indicators of the income generated by asset confiscation work, we assume this to be a function both of profitable criminality (driving the amount of assets available to be confiscated in each Member State based on a given amount of investment) and investment in policing (representing the latent apparatus which each Member State is able to bring to bear upon asset recovery work, as well as the level of commitment to combating organised crime). In addition, the overall size of Member State economies has an impact on their asset confiscation costs and revenues.

This simple logic is appropriate given that research has revealed what would otherwise be a catastrophic lack of appropriate data.  Had there been more time available, more investment could have been made in developing the suite of indicators and improving their reliability and validity.  We shall return to this issue in our final recommendations.

Because the logic model assumes that the maximal legislative option approximates recent measures in the UK, it must be adjusted (in the case of the maximal legislative option incorporating mutual recognition) to account for the impact of increased utilisation of mutual recognition instruments (which has not formed part of the UK’s approach).  The basis of such an adjustment—which takes the form of an adjustment factor coupled with a sensitivity analysis—is that options targeting mutual recognition will ease cross-border enforcement and, thus, raise the amounts recovered by each Member State.  The extent of this depends, however, upon the status quo ante.  In particular, Member States which rely upon non-conviction based orders stand to enjoy a greater impact because they currently struggle to enforce these orders overseas (whereas, for conviction-based orders, the potential advantages of mutual recognition relate mostly to speed and efficiency rather than the more fundamental question of whether an order will be enforced).  To account for this difference we double the adjustment factor for those Member States which currently rely upon non-conviction based orders to a significant extent.

Two other potentially relevant factors are:

· the extent to which  each Member State is a popular destination for organised crime profits (relevant because under FD 2006/783/JHA, the default position is that the executing Member State retains 50% of the value of the recovered asset); and

· the extent to which proceeds derived from crimes committed within each Member State tend to be retained at home, transferred elsewhere within the EU, or transferred outside the EU.

As there are no reliable data on these factors, we do not take them into account in our analysis. Nevertheless, as a matter of logic and in line with the microeconomic decision model for the rational criminal, we suggest that Member States with relatively efficient law enforcement systems are likely, all else equal, to experience greater flight of illicit assets to other States.  Conversely, Member States with less efficient systems will tend to have a greater inflow of criminal assets. In these countries, criminals are likely to seek ways to transfer their illegal (cash) wealth into the legal economy (Europol, 2009). An error factor in this prediction is the fact that in some countries organised criminal groups benefit from being culturally embedded in particular locations and from being a recognised ‘brand’ in the local economy.  Where this is the case, organised criminal gangs tend to invest large proportions of their profits in the local community thus reducing the level of capital flight that might otherwise be predicted.

Proxy indicators

The three basic components of the logic model (administrative efficiency, profitable criminality and investment in policing) demand internationally comparable proxy indicators.  The proxies employed are outlined in Table 1.4.  We reiterate that these indicators refer to a whole country (e.g. quality of institutions) or a whole institution in a country (e.g. tax collection efficiency), and are but proxies for narrower concepts specifically impacting on asset confiscation outputs for which EU-27 data is not available.

Table 20 Components of the model for generalising profitability, with proxy indicators

Component || Proxy || Reason for inclusion || Source

Profitable Criminality || Composite Rule of Law indicator (2007) || The Rule of Law indicator is a composite generated by the World Bank’s ‘Unobserved Components Model’.  In essence, it rescales 80 individual indicators used to create the composite Rule of Law indicator and places them in common units. It then constructs the composite measure as a weighted average of the underlying individual indicators. || World Bank

Costs of organised crime for business (2007) || Perceptions of businessmen about the costs imposed on their business by organised criminal groups is one of the best available indicator of the size of organised criminal business in a country. It is a proxy which more specific than the rule of law indicator, but it also neglecting important aspects of organised crime. Thus, the two indicators complement in each other in gauging the scope of profitable organised crime in a country. || WEF (2008)

Administrative efficiency || Wastefulness of government spending  (2007) || Inefficient government spending is approximated by perceptions of businessmen who are major consumers of public services thus well placed to formulate informed judgements. || WEF (2008)

Aggregate tax collection costs to net revenue collected (2007) || Asset Confiscation is a particular and highly complex and costly form of tax collection. We argue that those States whose taxation systems are less cost effective will tend to resist additional pressures to improve asset confiscation unless it can be demonstrated that asset confiscation regimes will always be profitable. || OECD (2009) 

Investment in Policing || Expenditure on public order and safety, PPP EUR per 100 000 citizens (2007) || General government expenditure on public order and safety as well as number of police officers indicate the available capacities and sophistication of law enforcement authorities for improved asset confiscation work. || Eurostat

Number of police officers per 100,000 citizens (2007) || Eurostat

These chosen proxy indicators reflect the constraints of this short project and suffice to provide headline results for immediate support of policy decisions, the list of indicators can easily be expanded and refined in the future in order to develop a more accurate model. Table 21 reports the data as collected from the original sources per table 20.  It should be noted that these data are for 2007, whereas the UK data is for the 2008/09 financial year.  This does not represent an important shortcoming in our analysis as the institutional and environmental factors underlying asset confiscation work such as wastefulness of government spending are stable over time (Kaufmann et al, 2005).

For the subsequent data analysis and predictions these indicators are standardised using Z-scores where the sample mean is 0 and standard deviation is 1.  Furthermore, the standardised values were transformed in order to eliminate negative values by adding to the absolute value of the lowest score plus 1 (to avoid adjusting by a factor of 0).  By this means, the variable values used for the analysis are rendered of comparable magnitude.  They are also made positive while fully preserving the relative variance represented by them.  This use of Z-scores eliminates bias caused by underlying variable distributions having different shapes. Table 21 reports two variables additional to the proxies listed in Table 20.  The first is a purchasing power parity (PPP) variable to account for differences in price levels of Member States relative to the EU-27 average (in 2007), which is necessary for estimating law enforcement costs in a comparable manner as labour and capital inputs have different prices in different Member States.  The second is a gross domestic product (GDP) variable which indicates the relative size of Member State economies.

Table 21 EU-27 country scores of drivers and barriers to enhanced asset confiscation work

MS || Rule of law; -2.5 = lawless, 2.5 = lawful (2007) || Cost of OC for business; 7 = significant, 1 = insignificant (2007) || Govt. spending; 7 = wasteful, 1 = efficient (2007) || Ratio of tax collection costs to net revenue (2007) || Public order spending per 100,000 citizens (EUR, PPP) (2007) || Police officers per 100,000 citizens (2007) || PPP within EU; EU27 average = 100% (2007) || GDP at market prices, €m (2007)

BE || 1.30 || 1.8 || 4.2 || 1.40 || 4836.9 || 360 || 108.3 || 335,085

BG || -0.08 || 4.2 || 5.2 || 1.29 || 2504.3 || 481 || 46.2 || 30,772

CZ || 0.87 || 2.1 || 5.2 || 1.25 || 4120.3 || 421 || 62.4 || 127,331

DK || 1.96 || 1.2 || 3 || 0.62 || 3079.5 || 193 || 137.4 || 227,534

DE || 1.70 || 1.8 || 3.9 || 0.78 || 4570.9 || 305 || 101.9 || 2,432,400

EE || 1.13 || 2 || 4.1 || 0.86 || 3532.0 || 242 || 73.1 || 15,828

IE || 1.73 || 1.7 || 4.2 || 0.79 || 5738.8 || 290 || 124.5 || 189,374

HE || 0.80 || 2 || 4.7 || 1.69 || 2679.4 || 454 || 90.7 || 225,540

ES || 1.08 || 1.4 || 3.9 || 0.65 || 4908.8 || 469 || 92.8 || 1,053,537

FR || 1.38 || 2.1 || 4.1 || 0.97 || 3526.6 || 371 || 108.1 || 1,895,284

IT || 0.40 || 4.4 || 5.8 || 1.16 || 4576.5 || 178 || 102.9 || 1,546,177

CY || 1.06 || 1.7 || 3.4 || 5.80 || 4881.8 || 647 || 88.1 || 15,951

LV || 0.73 || 1.8 || 4.8 || 1.31 || 3685.1 || 364 || 66.6 || 21,111

LT || 0.64 || 2.1 || 5 || 0.98* || 2423.5 || 334 || 60 || 28,577

LU || 1.76 || 1.5 || 3.4 || 1.18 || 5868.4 || 308 || 115.3 || 37,491

HU || 0.88 || 2.4 || 5.7 || 1.15 || 2981.8 || 263 || 66.7 || 100,742

MT || 1.58 || 1.3 || 4.1 || 0.97 || 2608.0 || 467 || 75.5 || 5,480

NL || 1.74 || 2.1 || 3.2 || 1.11 || 6160.6 || 218 || 101.9 || 571,773

AT || 1.93 || 1.4 || 3.7 || 0.64 || 4588.1 || 319 || 102.2 || 272,010

PL || 0.41 || 3.5 || 5.3 || 1.42 || 2406.4 || 258 || 62 || 311,002

PT || 1.01 || 1.5 || 4.4 || 1.41 || 2869.2 || 487 || 85.7 || 168,737

RO || -0.05 || 3 || 5.2 || 0.91 || 2193.2 || 211 || 63.8 || 124,729

SI || 0.89 || 2.1 || 4.7 || 0.83 || 3467.5 || 392 || 79 || 34,568

SK || 0.49 || 2.6 || 5.2 || 2.41 || 3172.4 || 261 || 63.2 || 54,905

FI || 1.86 || 1.3 || 2.9 || 0.77 || 3540.3 || 153 || 119.9 || 179,702

SE || 1.86 || 1.7 || 3.3 || 0.41 || 4231.7 || 193 || 115.7 || 337,944

UK || 1.66 || 2.7 || 4.7 || 1.10 || 7546.7 || 266 || 112.6 || 2,052,847

Equations

The predicted asset confiscation profits were derived by combining the theoretical model with the available cost and revenue estimates of the UK. This has been done by assuming that asset confiscation costs or revenues would surpass the UK’s costs and revenues if the respective net drivers and barriers scores in the given Member State exceed the UK values (e.g. if the given Member State has a more efficient tax collection system than that of the UK, it is expected to achieve the same asset confiscation revenue under lower costs ceteris paribus).  Comparison takes a linear multiplicative form, i.e. we assumed that drivers and barriers multiply each other’s impacts. This is justified by the fact that achieving revenue from asset confiscation work requires a series of institutions such as police, courts, financial investigators, etc. to function properly simultaneously.  For example, having an excellent judicial system in combination with zero investigative capacity will result in zero achievement.

By implication, the analytical model can be described by the following:

where MSi refers to the ith Member State and revenue and cost refer to asset confiscation work in financial year 2008/2009.

Revenue of the ith Member State is generated in the following way:

where Profitable criminalityMSi is the arithmetic average of the Composite Rule of Law (2007) and the Costs of organised crime for business (2007) indicators; and Investment in policingMSi is the arithmetic average of Expenditure on public order and safety (2007) and Number of police officer per 100,000 citizens (2007) indicators, GDP MSi is the gross domestic product at market prices in 2007.

Cost of the ith Member State is generated in the following way: 

where Administrative efficiency MSi is the arithmetic average of Wastefulness of government spending  (2007) and Aggregate tax collection costs to net revenue collected (2007) indicators. Simple arithmetic averages are used when aggregating constitutive indicators because we lack reliable knowledge about the relative importance of each indicator; arithmetic average assigns equal weights to each indicator.

Modelling outputs

These equations yield the predicted asset confiscation revenue, cost, and profit figures per member state as highlighted in table 22 and table 23.

Table 22 Predicted profits for the ‘maximal’ legislative option, EU27

Member State || Revenues (€m) || Costs (€m) || Profit (€m)

Spain || 131.00 || 36.57 || 94.43

UK || 188.08 || 131.08 || 57.00

Italy || 167.56 || 124.86 || 42.70

Poland || 31.66 || 10.87 || 20.79

Czech Republic || 19.21 || 4.27 || 14.94

Netherlands || 32.49 || 18.33 || 14.15

France || 105.42 || 91.92 || 13.49

Germany || 109.13 || 96.73 || 12.40

Greece || 19.24 || 13.80 || 5.44

Bulgaria || 10.78 || 5.77 || 5.01

Romania || 8.89 || 4.12 || 4.77

Portugal || 12.88 || 8.32 || 4.56

Hungary || 8.02 || 4.43 || 3.59

Slovakia || 5.85 || 2.38 || 3.47

Belgium || 21.39 || 19.22 || 2.17

Lithuania || 2.74 || 0.75 || 1.99

Slovenia || 3.43 || 1.45 || 1.98

Latvia || 2.37 || 0.76 || 1.61

Cyprus || 2.11 || 1.49 || 0.62

Estonia || 1.00 || 0.48 || 0.53

Malta || 0.26 || 0.18 || 0.08

Luxembourg || 1.48 || 1.51 || -0.03

Sweden || 7.67 || 8.62 || -0.96

Austria || 7.74 || 9.27 || -1.53

Ireland || 7.66 || 9.89 || -2.23

Finland || 1.99 || 4.33 || -2.35

Denmark || 1.73 || 5.79 || -4.06

Source: own calculation

Table 23 Distribution of predicted revenues, costs, and profits across EU-27 Member States, 2008/2009, million EUR

source: own calculation

As can be seen from the above table and graph, enhanced asset confiscation work would yield positive financial profits in all but 5 EU Member States. Due to imprecision of the data and the restrictive assumptions used to arrive at predictions, we recommend using a less refined scale categorising Member States into three broad groups (table 24). This categorisation summarises not only the absolute predicted profit per EU Member State, but also the predicted relative profitability of their efforts understood as the ratio of asset confiscation revenues and costs.

Table 24 Profitability of the ‘maximal’ legislative option, EU27

Member State || Categorisation || Profit ratio (profit/cost)

Czech Republic || highly profitable || 3.50

Lithuania || highly profitable || 2.63

Spain || highly profitable || 2.58

Latvia || highly profitable || 2.12

Poland || highly profitable || 1.91

Slovakia || highly profitable || 1.46

Slovenia || highly profitable || 1.37

Romania || highly profitable || 1.16

Estonia || highly profitable || 1.11

Bulgaria || moderately profitable || 0.87

Hungary || moderately profitable || 0.81

Netherlands || moderately profitable || 0.77

Portugal || moderately profitable || 0.55

Malta || moderately profitable || 0.46

UK || moderately profitable || 0.43

Cyprus || moderately profitable || 0.42

Greece || moderately profitable || 0.39

Italy || moderately profitable || 0.34

France || moderately profitable || 0.15

Germany || moderately profitable || 0.13

Belgium || moderately profitable || 0.11

Luxembourg || not profitable || -0.02

Sweden || not profitable || -0.11

Austria || not profitable || -0.17

Ireland || not profitable || -0.23

Finland || not profitable || -0.54

Denmark || not profitable || -0.70

Source: own calculation

The analysis thus far has not yet accounted for a crucial aspect of the maximal option which transcends national borders: increased utilisation of mutual recognition instruments. As this is a crucial aspect of the proposed set of policy options, we are bound to comment on this important issue.

There is, unfortunately, no data which would allow us to gauge the potential or actual magnitude of utilisation of mutual recognition either across the whole EU or per individual Member State.  We have therefore been forced to rely on a range of potential parameters in order to scope the magnitude of impact of mutual recognition on profitability. It is assumed that Member States where non-conviction based confiscation constitutes a considerable proportion of asset confiscation work can benefit relatively more (double the benefit derived by other Member States) from strengthened mutual recognition as their scope for alternative solutions is more limited currently. Countries which are considered as such are Bulgaria, Ireland, Italy, Romania, and the UK.

In the following sensitivity analysis, we assume that mutual recognition would increase the revenues of asset confiscation, but would also entail a modest increase in costs.  We assume that the former outweighs the latter in a 10:1 ratio, reflecting the fact that by the time assets have been traced to overseas locations, most of the investigative effort has been spent, permitting it to be treated as a sunk cost when analysing the marginal benefits brought by utilisation of mutual recognition instruments.  Parameters used in sensitivity analysis are shown in Table 25.

Table 25 Range of parameters used in the sensitivity analysis

Scenario || profit aspect || Predominantly conviction-based || Significant NCB element

5% || Revenue factor || 1.050 || 1.100

Cost factor || 1.005 || 1.010

10% || Revenue factor || 1.100 || 1.200

Cost factor || 1.010 || 1.020

15% || Revenue factor || 1.150 || 1.300

Cost factor || 1.015 || 1.030

20% || Revenue factor || 1.200 || 1.400

Cost factor || 1.020 || 1.040

25% || Revenue factor || 1.250 || 1.500

Cost factor || 1.025 || 1.050

30% || Revenue factor || 1.300 || 1.600

Cost factor || 1.030 || 1.060

50% || Revenue factor || 1.500 || 2.000

Cost factor || 1.050 || 1.100

Note: The cost factor is assumed to be 10% of the revenue factor and the non-conviction based regimes have double factors compared to the conviction based regimes.

The results of the sensitivity analysis show that the already profitable countries would benefit the most from utilisation of an efficient and effective system of mutual recognition. Results are less promising at the other end of the scale, however, four Member State become profitable upon consideration of mutual recognition impacts, viz:

· Luxemburg (5% scenario or higher);

· Sweden (15% scenario or higher);

· Ireland (20% scenario or higher); and

· Austria (25% scenario or higher).

Error! Reference source not found.6 presents revenue, cost and profit data adjusted to account for mutual recognition. The 5%, 15% and 50% scenarios are shown.

Table 26 Profitability for EU27, including mutual recognition  `

General points of interpretation regarding validity

Our profitability analyses, as already noted, presumes increasing utilisation over time, in the absence of which asset recovery work is likely to be less profitable.  However, whilst some policy actions are designed to directly raise utilisation, we do not necessarily regard these policy options as sufficient.  Rather, they represent steps which the EU is able to take in the context of the legal framework on the confiscation and recovery of criminal assets.  These mostly target political will, but there are also some important practical considerations which fall outside the scope of the EU legal framework.  These include the need for greater financial investigation capacity to generate system throughput (i.e. cases in the ‘pipeline’).  This requires training which, in turn, requires infrastructure and an up-front commitment of resources.  This, according to many practitioners with whom we spoke, represents a significant barrier to increased utilisation (especially in Eastern Europe, where there were, accordingly, questions raised about our policy action #17).  Expansion of financial investigation capacity is, however, not just a matter of quantity, but also quality.  In addition to resources, there is a knowledge input requirement, which presents as a significant barrier in Member States which do not yet have sophisticated financial investigation capacities.  Both of these areas are potentially apt for EU-level intervention in support of the options under consideration in this study.[62]

It can be argued pro tem that Member States in the highly or moderately profitable categories may directly benefit from adopting an enhanced asset confiscation system with a similar cost-benefit structure to the UK in 2008/2009.  However, this does not mean that Member States with different political-administrative systems cannot adopt different institutional solutions to become profitable or increase their profitability despite our findings.  In this regard, it should be reiterated that all of our proxy variables refer to country level or country institutional level characteristics, i.e. we could not employ asset confiscation specific indicators and thus could not take into account the specific characteristics of asset confiscation work compared to the wider law enforcement environment.

It should also be remembered that unprofitability in the narrow sense considered here does not imply that Member States will not achieve a net economic benefit once higher-order impacts are taken into account, to say nothing of the value of social and environmental benefits.

General points of interpretation regarding timing

The foregoing analysis centres on a point in time (April 2008 – March 2009) during which there is reasonable availability of UK cost data.  As already noted, this data reflects six years of effort to raise utilisation following the introduction of the Proceeds of Crime Act 2002, prior to which utilisation was minimal.. Several important points flow from this.

First, whereas asset recovery work in the UK was profitable in the year considered, it does not necessary follow that it was profitable in each of the preceding six years.  In fact, the opposite is likely to be true, of the UK as of most Member States, because results take time to manifest as cases progress through the system resulting, finally, in valuable assets vesting in the state.  Costs, on the other hand, will be more evenly distributed, as capacity is continually added; they may even be greater in the early stages, as new policies are implemented and, perhaps, new agencies stood up.

Second, the predicted impacts may take more or less time to manifest in other Member States than in the UK.  This is a question which concerns each aspect of a state’s asset confiscation apparatus, from time taken to increase financial investigation capacity, to the time taken for a case to progress through the judicial system, to the time taken for assets to then vest finally in the state.  As we are mainly focused here on the question of whether asset recovery work is potentially profitable, we do not make further adjustments to the data.

Third, the fact that utilisation was previously minimal is critical to properly interpreting the EU27 profitability results.  Essentially, it means that Member States can achieve the predicted results from a standing start.  By implication, Member States which have already begun to uplift utilisation, they could exceed the predicted outcome.

Fourth, the foregoing model, fixed as it is upon one point in time, does not predict what happens next.  Whilst it seems reasonable to assume that profitability does not suddenly decline, we have attempted to take account of uncertainty in developing a rough (minimum/maximum) prediction for future profits in the UK.  This is presented as Table 27, with the 2008/09 data for UK (converted to €) at year 6.  The projections thus represent what could happen in a country with an asset confiscation operation the same size as that in the UK; they can be scaled to account for different sized operations.  In any event, the uncertainty involved is so great that the projections are unlikely to assist policy-makers.

Table 27 Annual profit predictions for UK asset recovery work

Source: Authors’ calculations  performed in £ and converted to € using the 2009 average annual exchange rate from European Central Bank as of 11/8/2010, €1 = £0.891.

Specific points of interpretation

The proxy indicators upon which our results are based can now be used to interpret the results.  We focus here on those Member States which, according to the profitability model, are moderately profitable or unprofitable.  According to the data available and the theoretical model developed, the reasons for negative or moderate are essentially threefold. They are presented, together with country examples, in table 28. Error! Reference source not found.

Table 28 Main causes of negative profitability and potential remedies

Reasons for negative or moderate  profitability || Relevant indicator || Country examples || Potential remedy

low level of potential asset confiscation revenue

not enough assets to confiscate || · High level of rule of law · Low level of reported interference of organised crime || Finland, Denmark || · Not available

not enough capacity to confiscate || · Low number of police officers · Low spending on public order and safety || Slovenia,  Estonia, Lithuania Romania, Hungary || · Increase capacity by training or spending more · More efficient utilisation of available resources

high cost of asset confiscation

Questionable levels of efficiency in financially oriented law enforcement || · High level of wasteful government spending · Low efficiency of tax collection · || Belgium, Greece, Portugal || · Create specific organisations dedicated to asset confiscation work where efficiency is higher than average public service ·

Source: own categorisation

However, due to a number of restrictive assumptions, the narrow focus of the available proxies and the form of the theoretical model, some countries may do much better than predicted by this model.  For example, in Ireland, asset confiscation work by the specialist Criminal Assets Bureau (CAB) has proven to be profitable, partly due to a very effective utilisation of revenue powers (i.e. the ability to levy income tax owed against undeclared income in cases where its illicit origin cannot be proved).[63]  The work of CAB is, however, complemented by that of local police officers.  Because the capacity of the CAB is limited, there is a strong focus in Ireland on ‘mainstreaming’ local asset recovery work in this way, but the profitability of this component is not known.  Nevertheless, the Irish example shows that Member States may be able to establish profitable specialist agencies irrespective of the profitability of mainstream asset recovery work.[64]  On the other hand, the very promising results for Spain are difficult to reconcile with the opinion of one expert whom we interviewed that Spain’s (Napoleonic) system of enforcement through local courts makes it very difficult to recover assets cost effectively.  This may suggest that the proxy indicator for administrative efficiency takes too little account of enforcement mechanisms and/or efficiency within the judicial system.  This is an aspect of the model which warrants further research.

In any case, we draw attention to an appropriate way to interpret the foregoing rankings.  Non-profitability as estimated here flows from certain deficiencies in the Member State involved.  Whilst our profitability analysis may disregard many potentially relevant factors at (including any factors unique to particular Member States) this does not mean that Member States cannot benefit from attending to any deficiencies which factor in our analysis.

[1]               "An open and secure Europe serving and protecting the citizens", Council document 17024/09, adopted by the European Council on 10/11 December 2009.

[2]               Council document 7769/3/10.

[3]               COM(2010) 673 of 22.11.2010.

[4]               COM(2011) 307, 308 and 309 and C(2011) 3673 final of 6.6.2011.

[5]               COM(2011) 376 final of 24.6.2011.

[6]               "Proceeds of organised crime - Ensuring that 'crime does not pay'", COM (2008) 766 final.

[7]               Framework Service Contract No JLS/2010/EVAL/FW/001/A1, Study for an Impact Assessment on a proposal for a new legal framework on the confiscation and recovery of criminal assets.

[8]               From Austria, Bulgaria, Cyprus, Finland, France, Greece, Ireland, Italy, Luxembourg, Poland, Romania, Slovakia, Sweden, UK. CARIN is an international network of asset recovery practitioners which has Members in over 50 countries and jurisdictions. CARIN foresees one law enforcement and one judicial contact point per country

[9]               However, during the discussions some delegations expressed the wish to have more information on certain issues, such as non-conviction based confiscation.

[10]             Available at http://ec.europa.eu/home-affairs/policies/crime/crime_confiscation_en.htm

[11]             Report from the Commission pursuant to Article 6 of the Council Framework Decision of 24 February 2005 on Confiscation of Crime-related Proceeds, Instrumentalities and Property (2005/212/JHA), COM(2007) 805.

[12]             Report from the Commission based on Article 14 of the Council Framework Decision 2003/577/JHA of 22 July 2003 on the execution in the European Union of orders freezing property or evidence, COM(2008) 885 final.

[13]             Report from the Commission pursuant to Article 22 of the Council Framework Decision 2006/783/JHA of 6 October 2006 on the application of the principle of mutual recognition to confiscation orders, COM(2010) 428. 

[14]             Report from the Commission based on Article 8 of the Council Decision 2007/845/JHA of 6 December 2007 concerning cooperation between Asset Recovery Offices of the Member States in the field of tracing and identification of proceeds from, or other property related to, crime, COM(2011) 176 of 12 April 2011.

[15]             Such as the meetings of the CARIN Network or of the informal EU Asset Recovery Offices Platform.

[16]             For example the CEART Seminar and the Eurojust Strategic Seminar held in 2010.

[17]             For example the Commission services held several bilateral meetings with representatives of the FLARE (Freedom, Legality and Rights in Europe) Network and their associated networks

[18]             For example, substantial differences exist between the national regimes for third party confiscation and extended confiscation. Non-conviction based confiscation is heavily used in Ireland, the United Kingdom and other countries, but is not used in Romania (property is presumed to be of licit origin and confiscation based on a burden of proof of balance of probabilities may be perceived as problematic). France introduced a new crime for "possession of unjustified assets" (in case of evident links with organised crime activities) which does not exist in other Member States.

[19]             In Italy the proceeds of organised crime laundered  in 2011 were estimated at  € 150 billion (Bank of Italy, 2011).

[20]             For example in 2009 €189m were recovered in the UK and €60m in the Netherlands.

[21]             Home Office (2006), referred to in the 2010 Organised Crime Threat Assessment.

[22]             This figure is likely to underestimate the proportion of criminal wealth recovered, because i) it is net of expenses paid to private receivers, ii) it does not include amounts recovered in favour of victims and iii) for non-financial assets it records values realised at auction, which may be less than values reported stolen. Even so, the data suggests that the vast majority of criminal wealth goes unrecovered, especially given that the £15bn estimate relates only to organised crime. 

[23]             Respectively by the Bank of Italy and the Italian Court of Auditors.

[24]             Annex 2 contains some statistics on the value of assets recovered, but there is scarce data on organised criminal turnover against which to compare them.

[25]             Justice and Home Affairs Council Conclusions on confiscation and asset recovery of June 2010, Council document 7769/3/10. For similar statements see also the Executive Summary of the EU Organised Crime Threat Assessment 2011 and the Eurojust Annual Report 2010.

[26]             For example the Council of Europe Convention on laundering, search, seizure and confiscation of the proceeds of crime and on the financing of terrorism (CETS 198), which to date has been signed by 20 EU Member States and the European Union, and ratified by 12 Member States. The United Nations Convention against Corruption (UNCAC), which has been signed and ratified by almost all EU Member States and the European Union, has a Section on asset recovery.

[27]             Some EU legal provisions are not well coordinated. For example Framework Decision 2005/212/JHA establishes alternative criteria for extended confiscation as follows: i) A court is convinced that the property is derived from criminal activities of the convicted person prior to conviction; ii) A court is convinced that the property is derived from similar criminal activities of the convicted person prior to conviction; iii) The value of the property is disproportionate to the lawful income of the convicted person and the court is convinced that the property derives from criminal activity. The Framework Decision leaves Member States with the option to transpose one, two or all three criteria. This provision is not coordinated with the provisions on the grounds for refusal of mutual recognition of confiscation orders laid down in Framework Decision 2006/783/JHA. As a result the scope for mutual recognition of confiscation orders is restricted. The authorities in one Member State are obliged to execute confiscation orders issued by another Member State only if these orders are based on the same alternative criteria applied in the Member State receiving the order.    

[28]             Executive Summary of the EU Organised Crime Threat Assessment 2011.

[29]             The impact on third countries is not identified as positive or negative, as it is not known where assets would be moved or re-invested and an inflow of (laundered) criminal money may affect different countries (e.g. a developing country vs. a small country which is known as a tax haven) in a different way.

[30]             As stated above the fact that the EU legal framework is incompletely transposed is only part of the problem. It is also incorrectly transposed and, above all, insufficient to address the shortcomings identified in this impact assessment.

[31]             Which corresponds to recovery only from mala fide third parties that have paid less than market value.

[32]             The most recurring example is that of agricultural cooperatives.

[33]             The proposed option would introduce non conviction based confiscation measures and third party confiscation only in a limited way.

[34]             The underlying assumption is that Member State agents would not request the freezing of assets in other Member States (which is more costly and time-consuming than a national procedure) unless the value of the criminal assets identified and the chances of recovery are sufficiently high.

[35]             It is not yet clear whether an authority (such as the Ministry of Justice) will act as centralised national contact point for the data collection, nor whether the reporting requirement will also include a requirement to make the data publicly available.

[36]             Fieldwork interview.

[37]             This act provided for extended confiscation wherever a ‘course of criminal conduct’ was identified.

[38] PIU (2000), 5.

[39]             See part of POCA 2002.

[40]             http://library.npia.police.uk/docs/hmcpsi/AssetRecovery.pdf

[41]             This is especially interesting because, with the exception of drug trafficking, the number of acquisitive crimes recorded in the UK has fallen during the relevant period: see http://rds.homeoffice.gov.uk/rds/pdfs09/recorded-crime-2002-2009rev.xls.  There are many possible ways to reconcile the statistics, but there is no need to do so here.

[42]             Joint Thematic Review 2010, paragraph 2.13.

[43]             Prior to introducing the Proceeds of Crime Act 2002, the UK government did an estimate of implementation costs.  It has since maintained an interest in the costs and benefits of the legislation.  Profitability is one aspect of this.

[44]             Under ARIS, agencies receive 50% of amounts the recovery of which they are solely responsible for.  Where responsibility is shared this amount is apportioned.  For example, criminal confiscation pursuant to POCA section 6 involves contributions from the Police authorities (financial investigation), the CPS (obtaining confiscation orders following successful criminal proceedings), and HM Courts Service (enforcement), and these agencies receive, respectively, 18.75%, 18.75% and 12.5% of the revenue.

[45]             http://www.mpa.gov.uk/committees/finres/2010/100923/07/#h1000

[46]             Data from 2008/09 suggests that HMRC recovers through conviction-based confiscation and NCB cash seizure-forfeiture in a similar ratio to the police authorities.  This is important for their comparability, as the seizure-forfeiture regime, by virtue of its simplified procedure, is more profitable overall.  As regards conviction-based proceedings, those of the HMRC tend to be more complex and expensive to run, but they also tend to involve higher value proceeds (although these are often too well hidden to be recovered).

[47]             The results of this exercise have been tabled in parliament, see: http://www.publications.parliament.uk/pa/ld200809/ldhansrd/text/90720-wms0004.htm#column_WS163; SOCA’s statement of accounts is useful in interpreting these figures: www.official-documents.gov.uk/document/hc0809/hc08/0870/0870.pdf.

[48]             Recent legislative amendments have given the CPS the power to also bring civil confiscation proceedings, but these are yet to be exercised.

[49]             Thematic review, para 6.12.

[50]             Thematic review, para 6.12.

[51]             Much of this is done by the Department of Work and Pensions.  We understand from expert interviews that this work is likely to be no less profitable than that of other agencies.

[52]             There are also some costs borne by the Home Office (e.g. maintaining the JARD database) but these are negligible in the context of this analysis.

[53]             On the one hand, much of the evidence for confiscation cases has already been heard in the context of the criminal proceeding, and there are no jury costs.  On the other hand, these cases are sometimes heavily contested.

[54]             Ministry of Justice statistics.

[55]             http://www.nao.org.uk/idoc.ashx?docId=9db2d94f-0642-41f7-a697-334b2040ffdd&version=-1

[56]             In principle, given unlimited time and access to data held on JARD, it should be possible to reconstruct this pipeline for a more exact understanding of the system.  However, it is not necessary to do this in order to assess profitability, given that the UK situation is not unusual (more complex cases will tend to take longer to finally determine in all Member States), and also given that profitability is generally assessed with reference to financial-year accounts.

[57]             We test current year and one, two and three years previous. Tests do not find statistical significance with two- and three-year lags, possibly due to the limited time-series.

[58]             The purity of this deterrent logic is questioned by some defence lawyers in the UK and other Member States, on the basis that confiscation proceedings constitute an oppressive interference against a defendant’s capacity to defend criminal charges by diverting attention away from preparing a defence and into rearranging personal finances to deal with freezing orders.  The suggestion is that confiscation proceedings are sometimes instituted for tactical reasons related to the goal of prosecution.

[59]             This bias will not be as strong as one would expect from a rational decision-maker with perfect foresight: it is not always possible to know in advance which cases are ideal targets for investigation and prosecution.  Even where it is, these cases will not present themselves at

[60]             Paradoxically, there is evidence that some police forces may lay off financial investigators in the near future—even though their work is profitable in the narrow sense considered here—due to financial constraints.  The explanation is that ARIS returns only 50% of recovered revenues to frontline agencies, making it possible for asset recovery work which is profitably to the UK government overall to entail opportunity costs for the agencies involved, i.e. if it is not sufficiently profitable to be self-financed from ARIS receipts.

[61]             This reflects the actual increase in the number of persons trained as financial investigators (from 2283 to 2622) in the period from 2008/09 to 2009/10.

[62]             This is already acknowledged within the EU.  Consider, e.g., recommendation 4 in section 6.2 of the Council fifth round report on the UK: ‘the role and powers of financial investigators as well as their training system need to be presented at EU level and taken into account when common EU standards or training projects are being developed.’

[63]             Data from CAB annual reports. This data has not been fed into our profitability model for two reasons: because it does not represent the whole of the Irish asset confiscation apparatus (it only includes work done by CAB and even then it does not include court costs and enforcement costs), and because the Irish NCB system (administered by CAB) employs a 7-year lag between assets being confiscated and realised, making profitability at a given point in time difficult to estimate from on the available data.

[64]             The Netherlands is another example of a Member State which has established a profitable specialist agency: specifically, the Bureau Ontnemingswetgeving Openbaar Ministerie (BOOM), which is profitable according to the website of the Openbaar Ministerie [public prosecution] although figures to support this claim are not provided.