The missing link between “necessity” and “approximation of criminal sanctions” in the EU

Ever since the entry into force of the Amsterdam Treaty, the EU has had the competence to adjust national legislations via approximation instruments. In doing so, it has had the ability to influence the sanctions foreseen in the national criminal codes. The aim of this contribution is to review the scope of the EU’s competence to approximate national criminal sanctions with a view to assessing (1) the existence of an EU policy with respect to the approximation of sanctions and (2) the extent to which the legal basis has been used in a consistent and correct way. The frequently stressed sensitivity of the matter and the reluctance of Member States to give up their national sovereignty when it comes to criminal law, highlight the crucial need for a such policy and consistent and correct use of provisions. The contribution goes into the existing minimum maximum sanctions as well as possible future minimum minimum sanctions and maximum maximum sanctions.


I. Introduction
Ever since the entry into force of the Amsterdam Treaty, the EU has had the competence to adjust national legislations via approximation instruments. 2 In doing so, it has had the ability to influence the sanctions foreseen in the national criminal codes. 3 The Lisbon Treaty has not brought about significant changes with respect to the wording of the treaty approximation provisions 4 , as a result of which there is a certain continuity in the legal framework.
The aim of this contribution is to review the scope of the EU's competence to approximate national criminal sanctions with a view to assessing (1) the existence of an EU policy with respect to the approximation of sanctions and (2) the extent to which the legal basis has been used in a consistent and correct way. The frequently stressed sensitivity of the matter and the reluctance of Member States to give up their national sovereignty when it comes to criminal law 5 , highlight the crucial need for a such policy and make consistent and correct use of provisions. directives adopted in accordance with the ordinary legislative procedure, establish minimum rules concerning the definition of criminal offences and sanctions in the areas of particularly serious crime with a cross-border dimension resulting from the nature or impact of such offences or from a special need to combat them on a common basis.
The correct interpretation of those provisions proved to be challenging. Both the concept of "necessity" and the concept of "minimum rules" have been subject to complex legal discussions.

Necessity
Herlin-Karnell rightly argues that the necessity requirement for approximation is very vague which undoubtedly would give way to needless political discussions in the Council. 8 A solid interpretation of the necessity requirement presupposes that there is a clear understanding of the function of approximation and the goals that it is hoped will be achieved. Only if the function and goals are clear, is it possible to provide an argumentation justifying that in a specific case approximation is "necessary" to reach the goal. To this point in time, there is no common understanding on the functions of approximation. Various authors have commented on this. 9 Most commonly, a distinction is made between an autonomous and an auxiliary function for approximation.
The autonomous function for approximation is based on ideological, political and/or pragmatic motives and starts from the idea that a mature Area of Freedom, Security and Justice, includes a common view on the severity of crime and thus that sanctions that should be provided. 10 It is argued that differences in criminal sanctions would lead potential perpetrators to concentrate their criminal activity in the Member State with the mildest sanction regime. 11 A common view of the severity of certain crimes and the introduction of common minimum standards with respect to the sanction scales would in addition also be beneficial thereto for the deterrent effect of crime. 12 However, this autonomous function of approximation is not free from critique. On the occasion of the Council discussions on the draft framework decision on trafficking in human beings, the Danish delegation was not convinced of the effect approximation of sanctions could have and demanded a clear and stricter necessity test be carried out. 13 Rightly so, it was pointed out that the proposed approach to approximate sanctions via the adoption of minimum rules could never guarantee that after implementation a common view on the severity would be reflected in the criminal codes of the Member States. As an essential characteristic of approximation via the framework decisions, the requirements put to the Member States are only minimum requirements. Member States retain the discretion to maintain or even introduce new, more severe, sanctions. By definition, approximation via the introduction of minimum requirements will not result in a common view on severity reflected in the national criminal codes. Empirical analysis exists to support that argument. 14 As the criminal cultures of the Member States differ significantly, it should come as no surprise that the implementation of the minimum requirements has led to a wide variety of sanctions. Furthermore, the sanction included in the criminal code is only a small part of the sanction culture of a Member State. There are no internationally accepted standards on the links between in abstracto sanctions found in the criminal codes and in concreto sanctions imposed in a specific case. Even where in abstracto sanctions are the same, there is no guarantee that the in concreto sanctions will also be the same. In addition, there are also significant differences in the sanction execution legislation and practices of the Member States. 15 Looking to attain a common view on the severity of certain crimes requires a more holistic approach and should include far more initiatives to complement the introduction of minimum standards with respect to the in abstracto sanctions to be included in the national criminal codes. Claisse and Jamart add to that discussion that if the idea is to develop an Area of Freedom. Security and Justice in which there is a common understanding on the severity of crime, the current policy is very one-sided focussing only on sanctions involving the deprivation of liberty and not taking any initiative to then also approximate other sanctions such as financial penalties or alternative sanctions. 16 For all those reasons, the autonomous function is not convincing when looking into providing argumentation that would meet the necessity requirements. Owing to the inherent limitations of the 'minimum' character of the minimum requirements, any argumentation leading to the necessity of approximation to reflect the common understanding of the severity of crime such as would be suitable for a mature area of freedom, security and justice, fails to be convincing. As also pointed out by Nuotio, the supposed symbolic value of approximation is not the strong rational legal argument the European criminal policy maker needs to justify its interference in the sanction legislation of the Member States. 17 The auxiliary function of approximation starts from the provisions in the international cooperation instruments 18 , i. e. those provisions that make cooperation dependent on the sanctions that are nationally provided for. Previously, it was argued that double criminality requirements found in international cooperation instruments should be used as a basis to develop a solid necessity requirement for the approximation of the behavioural part of offences. 19 International cooperation is not only made dependent on the more frequently commented double criminality requirement, but is also made dependent on the severity of the offence underlying the cooperation. Therefore a parallel reasoning can be made with a view to developing a solid necessity requirement for the approximation of the sanctions for offences.
Given that it was deemed undesirable to make international cooperation possible for petty crime, it is provided for that cooperation is only possible for offences that meet a certain severity threshold. Owing to the growing importance of mutual recognition based cooperation instruments, some authors argue that the importance of the auxiliary function of approximation is declining or unconvincing. 20 However, that view is not entirely shared. Even in the more recent mutual recognition based cooperation instruments, sanction thresholds can be found. The effective use of international cooperation instruments is therefore dependent on the sanctions provided for in the national criminal codes. In contrast to the autonomous function of approximation, the auxiliary function can be a good basis to develop argumentation supporting the necessity of approximation of sanctions. In doing so, linking approximation of sanctions to the sanction thresholds found in cooperation instruments would be a good basis for a coherent/consistent European criminal policy in this respect. To ensure that international cooperation is possible, it is "necessary" to make sure that the offence meets the sanction threshold. The necessity of approximation would be to secure the possible use of a cooperation instrument. The fact that approximation is limited to minimum requirements and the unavoidable consequence that Member States may uphold or introduce a more sever sanction, in no way undermines the necessity argumentation.
Therefore, a critical analysis of the EU policy with respect to the use of the approximation competence requires a comparative analysis between sanction-related provisions in instruments governing international cooperation in criminal matters and the sanction related minimum requirements found in approximation instruments.

Minimum requirements
In addition to the discussion on the necessity requirement, the concept of "minimum requirements" has also been subject to discussion. To be able to interpret the meaning thereof, the Belgian national criminal code was used as a basis. In the Belgian criminal code, in abstracto sanctions include a minimum and a maximum sanction. 21 In doing so, a sentencing judge is provided with at times considerable discretion to impose what is deemed the most appropriate sanction in a particular case. Using the Belgian approach to include a minimum and a maximum sanction in the criminal code, the concept of 'minimum requirements' with respect to those sanctions, the possibility for the EU to adopt minimum requirements would be interpreted as minimum requirements with respect to the minimum sanction as well as minimum requirements with respect to the maximum sanction. This would result in an approximation instrument holding a minimum minimum sanction and/or a minimum maximum sanction. It should be noted though, that not all EU countries have a similar approach. The added value of minimum sanctions is often questioned. Not all EU countries have provided for minimum sanctions in their national criminal codes. 22 With a view to making sure that the specificities of the national criminal codes would be respected, a Declaration was added to the Amsterdam Treaty in which it was agreed that Member States that did not provide for minimum sanctions in their criminal codes could not be required to do so following the adoption of an approximation instrument. 23 This explains why the European criminal policy maker has focused the approximation efforts on the introduction of minimum maximum sanctions and none of the adopted approximation instruments hold provisions relating to minimum minimum sanctions. Interestingly, negotiators seem to have never brought up the declaration as a result of which it was not reincluded with the adoption of the Lisbon Treaty. This means that in the new legal 21 e.g. Article 468 Belgian Criminal Code: He who commits a robbery using force or threat, shall be punished with imprisonment from five years to ten years. 22  framework, the limitation with respect to the possibility to introduce minimum minimum sanctions has been lifted, which is particularly interesting given the lift on the unanimity requirement to formally adopt approximation instruments. An analysis of the EU sanction policy in a post-Lisbon era would therefore include both the current minimum maximum as well as the possible future minimum minimum sanctions.
For reasons of completeness, the question arises whether the current limitation of the EU's approximation competence would also allow the EU to adopt maximum requirements and in doing so provide standards with respect to either maximum minimum sanctions and/or maximum maximum sanctions. The line of argumentation provided for by Claisse and Jamart in this respect is worth mentioning. They argue that the treaty provisions should be interpreted in light of their spirit. The objective of the European policy maker, according to them, has been to allow the EU to adopt measures that would bring the criminal law provisions of the Member States closer together; put differently, to gradually reduce the diversity among the criminal law provisions and in doing so reflect a common view on the severity of certain crimes. 24 In light of the abovementioned discussion on the autonomous function of approximation and the inherent limitations to use approximation to reflect a common understanding of the severity of crimes, the need to complement the traditional limitation to minimum requirements with the possibility to introduce maximum requirements is justified. Bringing the criminal legislations closer together cannot be reached through the introduction of minimum requirements, but needs to be complemented with the introduction of maximum requirements. From a political perspective, the question arises whether the introduction of maximum requirements will ever be able to contribute to minimising the diversity amongst the Member States. The current criminal cultures are so diverse, that chances are that the political compromise reached in relation to the maximum requirements would be aligned based on the most punitive existing criminal provisions, leaving the other Member States with the discretion to introduce a more lenient sanction. This concern adds to the importance to link the introduction of requirements, be it minimum or maximum requirements, to the provisions found in international cooperation instruments. Even though the competence for the EU to introduce maximum requirements with respect to penalties can be contested on the strict reading of the treaty provisions, it remains interesting to assess to what extent a need can be deduced from the current cooperation provisions and to what extent the current treaty provisions providing the legal basis for approximation of sanctions need to be reviewed accordingly.
Taking account of the abovementioned extension of the EU's approximation competence and thus of the scope to not only include minimum requirements, but also maximum requirements, four types of approximation provisions can be anticipated in relation to: 1. minimum minimum sanctions; 2. maximum minimum sanctions; 3. minimum maximum sanctions; and 4. maximum maximum sanctions.
The functional analysis underlying this contributionstarting from the need to link approximation to the sanction-related provisions found in international cooperation instrumentshas led to the conclusion that only three out of those four options can be withheld.
Firstly, reference can be made to those provisions that ensure the proportionate use of cooperation instruments providing that the instrument cannot be used unless the maximum sanction provided for the underlying offence meets certain minimum standards. These provisions can be linked to the approximation of the minimum maximum sanctions.
Secondly, reference can be made to those provisions that ensure the proportionate use of cooperation instruments providing that the instrument cannot be used unless the sanction imposed for the underlying offences meets certain minimum standards. These provisions have been linked to the approximation of the minimum minimum sanctions. However, it should be added that the idea that the introduction of minimum minimum sanctions guarantees that the sanctions imposed do not go beyond that, fails to take account of the difference between in abstracto sanctions and in concreto sanctions. It is not uncommon for legal systems to have provisions that provide the sentencing judge with the discretion to impose a sanction that goes beyond the minimum provided for in the criminal code. Notwithstanding this reservation, it remains interesting to look into the links between sanction related cooperation provisions and minimum minimum approximation, especially given the recent proposals launched by the European Commission to that end.
Thirdly, the instruments governing the transfer of execution of sanctions, hold socalled adaptation provisions. Whenever the sanction imposed by one Member State is inconsistent with the law of the executing Member State, the latter is allowed to adapt the sanction. The executing Member State is allowed to adjust the foreign sanction in light of the maximum sanction it has provided for that offence in its own national criminal code. Those provisions can be linked to the possibility to avoid the need for adaptation via the introduction of maximum maximum sanctions.
Because the analysis failed to reveal a link between the introduction of maximum minimum sanctions and the added value it would have to facilitate or safeguard the use of international cooperation instruments, that combination is not further elaborated on. After all, without a link to an international cooperation instrument, it would be impossible to substantiate the need for approximation.

II. Minimum maximum penalties
A first possible type of approximation consists of introducing minimum requirements with respect to the maximum penalties. The necessity justification would consists of a reference to the severity thresholds found in international cooperation instruments: for international cooperation to be possible, it is necessary that the underlying offence meets the severity threshold and in absence thereof, it is necessary for the EU to take an initiative to ensure the possibility to cooperate and to that end introduce minimum requirements for the maximum sanction. The analysis first looked into the sanction-related cooperation provisions before reviewing approximation from an auxiliary perspective.

Sanction-related cooperation provisions
The majority of sanction-related provisions found in international cooperation instruments relate to minimum requirements with respect to the maximum sanction that is provided for in the national criminal codes. To be able to conduct an analysis of the extent to which the European policy maker takes account of those sanction thresholds when preparing approximation initiatives, it is important to identify those sanction-related cooperation provisions.
A maximum sanction of … … is required in the following provisions Missing link between "necessity" and "approximation of criminal sanctions" The analysis of the existing cooperation instruments reveals that a maximum sanction of 6 months is sufficient to ensure the possibility to conduct a house search and seizure using the provisions in the Schengen Implementation Convention. The aim to ensure the use of those provisions in relation to the offence that is subject to approximation is sufficient to successfully argue that there is a need to ensure that all Member States have provided for a maximum sanction that entails at least 6 months of imprisonment and in doing so meet the necessity requirement found in the approximation provisions. The extradition provisions from the Council of Europe Cybercrime convention, require that the underlying offence is punishable with imprisonment of at least one year. The aim to ensure the use of those provisions in relation to the offence that is subject to approximation is sufficient to successfully argue that there is a need to ensure that all Member States have provided for a maximum sanction that entails at least 1 year of imprisonment and in doing so meet the necessity requirement found in the approximation provisions. Obtaining information on bank accounts as governed by the provisions of the Protocol to the EU mutual legal assistance convention, is only possible if the underlying offence is punishable with a maximum imprisonment of at least 4 years. The aim to ensure the use of those provisions in relation to the offence that is subject to approximation is sufficient to successfully argue that there is a need to ensure that all Member States have provided for a maximum sanction that entails at least 4 years of imprisonment and in doing so meet the necessity requirement found in the approximation provisions. The overview of the sanction-related requirements found in the international cooperation instruments is the basis for the review of the current approximation instruments.

Auxiliary approximation
Since the introduction of the possibility thereto with the Amsterdam Treaty about 20 approximation instruments have been adopted. 25 The minimum maximum sanctions included therein vary from a maximum imprisonment of at least 1 year in the Directive on sexual exploitation of children 26 to no less than 15 years in the Framework Decision on terrorism 27 . Exceptionally, those minimum requirements are linked to the sanction thresholds found in cooperation instruments. In the Framework Decision on the forgery of non-cash means of payment and the effective, proportionate and dissuasive, but also allows for the persons involved to be extradited. 28 Similarly, in the more recent Directive on Trafficking in Human Beings, reference is made to effective, proportionate and dissuasive sanctions that allow for the persons involved to be surrendered. 29 The vast diversity in the minimum requirements found in the approximation instruments and the very spare references to cooperation requirements seems to support the idea that the European criminal policy maker lacks a comprehensive plan to ensure consistency in the development of the European sanction policy.
Importantly, an evaluation of the minimum requirements included in the approximation instruments needs to take account of the conclusions adopted by the JHA Council to work with sanction scales to support its approximation policy. Already, on the occasion of the political discussions leading up to the adoption of the first approximation instruments, concerns were voiced relating to the sanction requirements included therein. The proposal for a Framework Decision on trafficking in human beingsmirroring the sanction requirement included in the pre-existing Framework Decision on Fraud against the Europrovided for a minimum maximum sanction of 8 years. 30 In particular, the Danish delegation was not convinced of the added value of such a requirement. During the 28 -29 May 2001 JHA Council discussions on that proposed Framework Decision, the delegation rightfully argued that the mere fact that trafficking in human beings was punishable with a maximum imprisonment of 6 years in one Member State and 8 years in another Member State would have little to no effect on the cooperation between the two. 31 In light of the identified sanction thresholds included in the table abovewhich includes far more instruments then those applicable at the time of the Danish argumentthat position should be supported. The introduction of a minimum maximum imprisonment of 8 years has no added value whatsoever in relation to the sanction thresholds included in cooperation instruments. As a result of this discussion in the JHA Council, COREPER was tasked to analyse the appropriate severity of the minimum maximum sanction to be included in the Framework Decision on trafficking in human begins. In the context of that discussion Denmark argued that it was in the best interest of consistency in the EU policy to work with as little sanction scales as possible. Despite having built their argumentation in the initial JHA Council upon the limited added value of a minimum maximum sanction of 8 years in light of the effect it would have on cooperation, only one of the three sanction scales suggested by the Danish delegation is directly linked to coopera-

EuCLR
Missing link between "necessity" and "approximation of criminal sanctions" tion. 32 Their first category consists of sanctions that could lead to extradition, which at the time entailed an implicit reference to Art. 2.1. of the Council of Europe Extradition Treaty 33 , i. e. a minimum maximum imprisonment of 1 year and could be reinterpreted in light of Art. 4.1 of the Extradition Treaty between the EU and the US 34 to mean a minimum maximum imprisonment of more than 1 year. The proposal of the United Kingdom in its turn makes no link whatsoever to the cooperation instruments introducing a set of mere numerical categories. The compromise text prepared by the Belgian presidency retains the Danish proposal to link the first category to extradition and introduces three numerical categories based on the UK proposal. 35 Unfortunately, the text finally adopted during the 25-26 April 2002 JHA Council has dropped the reference to the extradition requirements and lists four sanction scales to be used as a basis for the approximation of sanctions: (1) sanctions between 1-3 years, (2) sanctions between 2-5 years, (3) sanctions between 5-10 years and (4) sanctions above 10 years. The choice for sanction scales as opposed to fixed sanctions is motivated by the desirability to allow Member States some flexibility when implementing the approximation provisionsgiven the diversity in the punitivity of the criminal cultures of the Member States. In doing so, each Member State couldin light of the punitivity of its own criminal culture choose the most suitable sanction based on the particular scale.
In light of the sanction-related cooperation provisions identified above, only the first two scales can be linked to cooperation requirements. The introduction of a minimum maximum sanction that exceeds 5 years imprisonment can in no way be justified in light of the added value for cooperation. Taking account of the then status of the political negotiations related to the European Arrest Warrant and the stability of the provisions relating to the abandonment of the double criminality requirement which is linked to a minimum maximum sanction of 3 years, 36 it is unfortunate that that a threshold of which the future importance could easily have been anticipated to at the time, was not reflected in sanction scales. Having adapted the sanction scale to refer to sanctions between 3-5 years would have been useful and would not have taken away from the idea to work with flexible sanction scales.
The Framework Decisions adopted after the JHA Council introducing the sanction scales, consistently refer to one or more of those sanction scales. Only for severe drug related crime, the need was felt to introduce a minimum maximum sanction that exceeds 10 year and in doing so falls within the 4th sanction scale. 37  majority of sanction provisions fall within the second sanction scale. Notwithstanding the consistency of the sanction provisions in light of the adopted sanction scales, the sanction policy must be criticized for lacking a solid link with cooperation and in doing so having a limited added value for cooperation making it hard to justify the necessity of approximation.
The fact that the Directives adopted upon the entry into force of the Lisbon Treaty do no perfectly mirror the sanction scales and use fixed numerical requirements 38 , is by some considered to be an example of incoherent policy making. 39 That position cannot be fully supported. Because the minimum maximum requirements by definition leave the Member States with a margin of appreciation and allow for a sanction to be introduced in the national criminal code that goes beyond the upper limit of the sanction scales, de facto there is no legal difference between the requirement to introduce a maximum sanction of at least 1 year and the requirement to introduce a maximum sanction of at least between 1 and 3 years. Providing the obligation for Member States to introduce a maximum sanction of at least 1 year does not limit the flexibility that was originally deemed of such importance that sanction scales should have preference over fixed sanction levels. Furthermore, the comment that requiring Member States to have a minimum maximum sanction of 3 years limits the degree of flexibility originally intended is justified. Linking the minimum maximum sanction of 3 years to the first sanction scale entailing a sanction between 1 and 3 years, would result in a loss of the possibility of introducing a maximum sanction of 1 or 2 years. Linking the minimum maximum sanction of 3 years to the second sanction scale entailing a sanction between 2 and 5 years, would result in a loss of the possibility to introduce a sanction of 3 years. However, despite the inherent loss of flexibility and the inconsistency in light of what was decided at the 25-26 April 2002 JHA Council, an evolution towards requiring a minimum maximum sanction of 3 years should be applauded in light of the sanction thresholds found in cooperation instruments, 40 regardless of the fact that that link was not recognised in the process of preparing the approximation instrument. All mutual recognition based cooperation instruments use this sanction requirement to limit cooperation without double criminality. In light of this, the importance of the 3-year-threshold will only increase in the future.
To conclude, there is far more potential in the adoption of minimum maximum sanctions then currently included in the necessity justification. Precisely because of the continued diversity in the national sanction legislation, especially after having implemented the approximation instruments and owing to the necessity require-

EuCLR
Missing link between "necessity" and "approximation of criminal sanctions" ment that is inherently linked to approximation, it should get preference to fully align future approximation with cooperation requirements. The political negotiations leading up to the adoption of the next approximation instrument should not focus on a compromise sanction requirement in light of a comparative legal analysis of the different Member State criminal codes, but should provide an answer to the question as to which cooperation instruments should be able to be used. The selection of those cooperation instruments will determine the sanction requirement to be included in the approximation instrument.

III. Minimum minimum sanctions
International cooperation instruments not only refer to minimum maximum sanctions as included in the national criminal codes. Besides those in abstracto sanctions, some cooperation provisions also make reference to in concreto imposed sanctions or to the in concreto remaining sanction. A second possible type of necessary approximation of sanctions therefore relates to the minimum minimum sanctions. In parallel to the reasoning related to the minimum requirements for the maximum sanctions, the necessity justification links in with the requirements related to the minimum imposed sanction. To ensure that cooperation is/remains possible, the imposed sanction should meet the minimum requirements included in the cooperation instruments. To that end, it may be necessary to introduce minimum requirements at EU level. It is fundamental however, to take account of the difference between minimum sanctions in abstracto provided for in the criminal codes and the sanctions actually in concreto imposed. Having included in the national criminal code a minimum sanction that mirrors the minimum requirement in the cooperation instruments is by no means a guarantee that the imposed sanctions will not go beneath that minimum. The European Commission rightly argues in its 2004 Green paper 41 that various legal techniques exist in the national legal systems to allow the sentencing judge to impose a sanction below the minimum sanction provided for in the national criminal code. Reference can be made to attempted crimes 42 , forms of participation to a crime 43 or other mitigating circumstances. 44 Despite the fact that the European Commission demonstrated awareness of this complexity, it is not sufficiently taken into account in the mirroring approximation initiatives. In analogy to the previous section, firstly the relevant cooperation provisions are identified and secondly the mirroring approximation initiatives will be reviewed. 43 Article 66 and 67 of the Belgian Criminal Code make a distinction between an accessory and an accomplice. The latter is considered to be a full co-author of the offence, whereas the former is sanctioned with a lower penalty. 44 Article 411 of the Belgian Criminal Code stipulates that manslaughter can be mitigated if it was immediately provoked by sever acts of violence.

Sanction-related cooperation provisions
Whereas most sanction-related cooperation provisions relate to the minimum maximum sanction included in the national criminal code, there are various cooperation instruments that have limited their application scope referring to the minimum sanction imposed in a specific case. To be able to assess the consistency in the European approximation policy, it is necessary to first identify and classify those provisions.
A (remaining) imposed sanction of … … is required in the following provisions of cooperation instruments: The analysis of the most important current cooperation instruments reveals that a European Arrest Warrant can only be issued provided that a penalty of at least 4 months was imposed. The aim to ensure the use of those provisions in relation to the offence that is subject to approximation is sufficient to successfully argue that there is a need to ensure that all Member States have provided for a sanction that entails that at least 4 months of imprisonment is imposed and in doing so meet the necessity requirement found in the approximation provisions. The transfer of the execution of a sentence involving the deprivation of liberty requires that the remaining penalty is at least 6 months. The aim to ensure the use of those provisions in relation to the offence that is subject to approximation is sufficient to successfully argue that there is a need to ensure that all Member States have provided for a sanction that entails that at least 6 months of imprisonment is imposed and in doing so meet the necessity requirement found in the approximation provisions.

Auxiliary approximation
Following the existence of the 8th Declaration to the Amsterdam Treaty in which it was provided that no Member State would be obliged to introduce a minimum sanction in case the national criminal system did not provide for minimum sanctions, the European Commission never took the initiative to explore the EuCLR Missing link between "necessity" and "approximation of criminal sanctions" introduction of minimum minimum sanctions in its approximation instruments. In old first pillar instruments however, examples of minimum minimum sanctions can be found. Reference can be made to directive 2001/51/EC on carrier liability in the Schengen area. In Article 4.1. b), it is stipulated that the minimum fine shall not be less then #3000. Following the disappearance of the 8th Declaration following the entry into force of the Lisbon Treaty, the European Commission had taken the initiative in two recent case files to propose the inclusion of minimum minimum sanctions. Firstly, reference can be made to the directive related to fraud affecting the financial interests of the Union. On 11 July 2012, the European Commission launched its proposal for a directive in which Article 8 included not only minimum maximum sanctions, but also a minimum minimum sanction of 6 months. To justify the introduction thereof, the European Commission argued that this approach would not only ensure that similar sanctions would be imposed in the different Member States and the deterrent effect of the sanction would be increased, but also that the introduction of this kind of minimum sanction would ensure the use of the European Arrest Warrant (2012:10). With respect to that first line of argumentation, it was already argued that similarity in the in abstracto sanctions by no means guarantees similarity in the in concreto imposed sanctions. With respect to the second line of argumentation, there is no empirical evidence to suggest that similarity in the in abstracto sanctions would impact in the deterrent effect. On the contrary, deterrence research has demonstrated that sanction provisions in general have very little deterrent effect and opportunity and arrest probability are far more important. 45 Especially the link with the European Arrest Warrant is interesting. As presented in the table included above, Art. 2.2. of the Framework Decision on the European Arrest Warrant refers to a minimum imposed sanction of 4 months. Even where a reference to the European Arrest Warrant is applaudable, it is most unfortunate that no legally sound solution was found. Arguing that setting the minimum at 6 months would on average lead to a conviction of no less than 4 months 46 is a weak necessity justification that could not convince the members of the Council. The European Parliament had tried to find a compromise and suggested the minimum minimum requirement to be lowered from 6 to 3 months, which would render the provision useless in light of the cooperation requirements. As the Council cannot be convinced, the JHA Council of 6 June 2013, agreed on a general approach for the directive on the protection of the EU's financial interest in which the reference to minimum minimum sanctions was dropped. Secondly, reference can be made to the directive on the protection of the Euro against counterfeiting, for which the European Commission launched a proposal a couple of months before the general approach on the above mentioned case file was reached. In analogy to the developments in the other case file, the Commission launched a proposal on 7 February 2013, in which it included minimum minimum sanctions of 6 months. To justify that provision, not only a reference to the European Arrest Warrant is included, but also to the cross border execution of sanctions. An analysis of the sanction-related cooperation provisions included in the table above, reveals that indeed the execution of foreign sanctions requires a remaining penalty of at least 6 months. Making sure that the imposed sanctions meet the 6 month requirement is a way to ensure that the execution of the imposed sanctions will be able to be transferred.
In light of the general approach reached in the other case file, the European Parliament was of the opinion that also in this case file the reference to minimum minimum sanctions could best be dropped. 47 Here too, the European Commission did not succeed in building solid and convincing argumentation on the need for minimum minimum sanctions as a result of which the 7-8 October 2013 JHA Council again reached a general approach from which the minimum minimum sanctions were lifted. Despite the improvement in the wording of the justification compared to what was included in the proposed directive on fraud affecting the Union's budget, the wording still does not take due account of the difference between in abstracto and in concreto sanctions. To ensure that transfer of execution cannot be refused based on the length of the remaining sanction, the minimum requirement should be related to the sanction to be imposed by the judge rather than the sanction to be included in the criminal code. The question arises whether the competence provisions in Art. 83 TFEU can be interpreted in a way to also include minimum requirements with respect to the imposed sanctions and whether it is acceptable to also limit the discretionary powers of the sentencing judge in light of what is needed to ensure the use of the international cooperation instruments. Given that the competence provisions refer to 'minimum rules with respect to sanctions', there is no technical-legal reason to inhibit an interpretation that also includes minimum rules with respect to imposed sanctions. The European Commission may have succeeded in convincing the JHA Council should it have found a way to demonstrate that in practice it happens that even for the EU's priority offences, the imposed sanctions are too low to guarantee possible cooperation. To be able to elaborate on such an argument, it is possible to conduct a comparative analysis on the sanctions imposed in the different Member States. It is well known however, that a reliable comparative statistical analysis is far from self-evident. 48 It is rather challenging not to only single out the convictions that relate to the EU's priority offences (and not to similar national offences) but also to get hold of information on the sanction that was imposed. Despite the fact that the European Commission recently had to succumb to the strong headed members of the JHA Council, the idea to introduce minimum minimum sanctions had significant potential if due account is taken of the caveats elaborated on above.

IV. Maximum maximum sanctions
A third possible type of approximation of sanctions in the EU consists of introducing maxima with respect to the maximum sanction. The introduction of maximum maximum sanctions is also referred to as "restrictive approximation". Vogel has pointed to the fact that the reason to engage in introducing maximum maximum sanctions is to avoid sanctions that are contrary to fundamental human rights principles. 49 In that respect, the abolishment of the death penalty via the sixth protocol to the ECHR could be seen as an example of such restrictive approximation initiatives. van Zyl Smit and Ashworth have built a similar reasoning, pointing to the fact that disproportionate sanctions could constitute a human rights violation. 50 In the 2004 Green Paper, the European Commission expressed its concern about the fact that in the current approximation practice, no initiatives on maximum maximum sanctions are being developed because the differences between the Member States can be very significant. Comparative legal analyses in the impact assessments related to the European legal initiatives indeed point to such extreme diversity. It is well known that in some countries, amongst which Belgium and the United Kingdom, life sentences are known, whereas both Portugal and Spain have constitutional issues with that. In relation thereto, the European Commission has questioned the use of life sentences in the past. 51 The suggestion has been put forwardstarting from the reintegration objectiveto replace life sentences with fixed terms in prison. It was argued that a person's behaviour in prison can only change for the better, provided that the person is given hope of being reintegrated into society. In doing so, the European Commission aligns itself with the Spanish argument, as life sentences would be incompatible with the mandatory task of prison authorities to work towards the rehabilitation of its prisoners. 52 From the above, it can be deduced that discussions on whether or not to introduce maximum maximum sanctions have been limited to the so-called autonomous function of approximation. With a view to reducing the difference between the criminal laws of the Member States, some consider it advisable to also work with maximum maximum sanctions. The analysis underlying this contribution starts from the auxiliary function of approximation and seeks to link any approximation initiative to an established need to render international cooperation in criminal matters possible, or at least facilitate it.
When looking into the position of imposed sanctions in the international cooperation mechanism, it becomes clear that not only the minimum of an imposed sanction may be important but that the maximum of the imposed sanction can also have a significant role to play. As pointed to in the introduction however, the EU's competence to engage in the introduction of maximum maximum sanctions can be questioned. Nevertheless, it remains interesting to look into the possible added value that maximum approximation could have for international cooperation in criminal matters to be able to look into the need to either broadly interpret the current competence provisions or seek to extend the current competence provisions to also include maximum approximation.

Sanction-related cooperation provisions
Even though international cooperation in criminal matters is never made dependent of the level of the maximum sanction, the diversity in maximum sanctions does create problems when engaging in cross-border execution of sanctions. Precisely because of the differences in the maximum sanctions, an adaptation mechanism was included in the framework decision on mutual recognition of sanctions involving deprivation of liberty. Art. 8.2. stipulates that should the duration of the imposed sanction be incompatible with the law of the executing Member State, the latter may adapt the sanction. 53 The adapted sanction cannot be lower than the maximum sanction provided for in the criminal code of the executing Member State. Similar adaptation provisions can also be found in other instruments governing cross-border execution of sanctions. Those adaptation provisions are not free of criticism, however. Firstly, the question arises as to how the provision relates to the abandonment of the double criminality requirement. As a fundamental feature of mutual recognition based cooperation, double criminality is abandoned for a list of 32 offences to the extent they are punishable in the issuing Member State with at least a 3-year imprisonment. 54 Reading both provisions together may lead to the interesting conclusion that despite the fact the execution of a foreign sanction imposed for an offence that is not punished in the executing Member State and therefore cannot be de iure refused for not meeting the double criminality standard, could de facto be refused by lowering the imposed sanction to the nationally foreseen maximum, whichin absence of an incriminationwould logically have to result in the omission of the sanction altogether. That conclusion would however not be compatible with the mutual recognition ideas behind the said framework decision. Chances are high the certificate for cross-border execution of the sentence would be withdrawn, which would lead to the undesirable situation where a person is deprived of the change to serve a foreign sentence in a prison in his home state. In doing so, the rehabilitation idea underlying the transfer of prisoners 55 to their home states is undermined.
Secondly, the question arises of how to go about an adaptation. Identifying the sanction that is nationally foreseen for the offences underlying the foreign conviction, pre-supposes that sufficiently detailed information on the offence is available. That supposition may be incorrect in practice. Additionally, referring to the maximum sanction provided for in the criminal code may not be sufficient. Mirroring the comment that the minimum sanction included in the criminal code by no means reflects the actual minimum a judge may impose in a specific case, the maximum sanction included in the criminal code also fails to reflect the actual maximum a judge may impose in any given case. Mechanisms such as concurrence, recidivism or aggravating circumstances may result in the possibility for a judge to impose a sanction that goes beyond the maximum sanction included in the criminal code. Limiting the adaptation process to the maximum sanction included in the criminal code would therefore have the undesired effect that the aggravating effect of a concurrence, recidivism or another aggravating circumstance is lost in the course of the cross-border execution of a sanction. This undesired effect can only be avoided by reviewing each individual case file in detail, which in no way is compatible with the original mutual recognition idea.
In light of this discussion, the question is raised as to what extent the introduction of maximum maximum sanctions should be considered as a future policy option in order to avoid the need, originally felt, to have an adaptation mechanism in the first place.

Auxiliary approximation
According to Miklau there is an imbalance in the current European approximation policy because there have been no successful initiatives to also look into the approximation of maximum sanctions. 56 The introduction of mere minimum maximum sanctions has not been able to reduce the diversity in the Member States. Precisely because of the fact minimum requirements are only a minimum, Member States have considerably made use of the possibility to introduce or maintain a more severe criminal policy. Elholm has illustrated this in referring to the Danish implementation of the Framework Decision of 13 June 2002 on Terrorism. 57 The Frame-work Decision requires that offences committed with a terrorist intent are punished more severely than offences with a different intent. 58 Whereas the basic nonterrorist offence was originally punishable with a 6-year prison sentence, when implementing the Framework Decision, Denmark decided to introduce a possible life sentence for offences committed with a terrorist intent. In doing so, a significantly higher sanction was made possible. It can be questioned however, as to what extent that sanction is compatible with the intentions of the European Commission, who in the past had already expressed concerns related to life imprisonment 59 and certainly would not want to be linked to the introduction of even more cases where life sentences are possible.
Interestingly, the existence of life sentences seem to have been the main reason for the introduction of an adaptation possibility in the instruments governing crossborder execution of sanctions. The abolishment of the life sentence possibly introduced with a number of maximum maximum sanctions could lift the need to have an adaptation mechanism in the first place. An analysis into this policy option should take due account of the inherent maximum character of the maximum maximum sanctions and therefore the possibility of the Member States to introduce or maintain a sanctions that is less severe. Furthermore, it should take account of the scope limitation of approximation in terms of the offences that can be subject to approximation which would not only impact on the action range of the EU to reduce the need to have an adaptation mechanism, but would also impact on the internal consistency of the national criminal codes (in which life sentences would be abolished for the more severe EU priority crimes, but would still be able to be maintained for less severe national crimes).

V. Conclusion
The analysis aimed at mapping the scope of the EU's approximation competence, elaborating on the possibility to interfere in the national sanction provisions and assessing to what extent that competence is consistently used.
Taking account of the formulation of the competence provisions in the EU Treaty and the sanction-related provisions found in international cooperation instruments, three possible types of approximation were identified.
Firstly, it was assessed as to what extent minimum maximum sanctions could support cooperation. Linking in with the sanction-related cooperation provisions, strong and convincing necessity argumentation can be built. Unfortunately, besides a couple of exceptions, currently no links can be found between approximation and cooperation instruments.

EuCLR
Missing link between "necessity" and "approximation of criminal sanctions" double criminality requirement are really important. Given that it is next to impossible to externalise a common vision on severity of crimes through the introduction of minimum requirements with respect to the maximum sanctions, it is highly recommended to link the necessity debate to the requirements in international cooperation instruments. Secondly, it was assessed as to what extent minimum minimum sanctions could support cooperation. Now that the 8th declaration to the Amsterdam Treaty is no longer valid, there is no doubt that the EU can take the initiative to explore the introduction of minimum requirements with respect to the minimum sanctions. Despite the fact that the European Commission recently had to succumb to the strong headed members of the JHA Council, the idea to introduce minimum minimum sanctions had significant potential if due account is taken of the caveats elaborated on above. In light of this, it is interesting that the European Commission has requested a study to be carried out looking precisely into those minimum sanctions, including an empirical phase in which statistical information will be compiled from all 28-EU Member States with respect to the sanctions that have been imposed for the EU's priority offences. The report is expected to be delivered early on in 2015. The debate on the levels of the minimum minimum sanction should be built around the requirements found in cooperation instruments.
Thirdly, it was assessed as to what extent maximum maximum sanctions could support cooperation. Not only is the EU competence to engage in the introduction of maximum maximum sanctions questionable from a strict reading of the treaty provisions, the elaboration of approximation initiatives that could support international cooperation in criminal matters by lifting the need for adaptation whilst not causing significant inconsistencies in the national criminal codes, requires an indepth complex legal analysis of the underlying mechanisms. Profound comparative legal research is necessary.