March 30
Anjum Shabbir
Anjum Shabbir
20th March 2020
Justice & Litigation

Analysis: “Non-conviction-based confiscation under EU Criminal Law: the CJEU and the “AGRO IN 2000” case” by Angelo Marletta

On 19 March 2020, the Court of Justice of the European Union delivered its judgment in the case ‘AGRO in 2000’ (C‑234/18), ruling that EU law does not preclude Member States from enacting forms of non-conviction based confiscation under national law.

The case originated from a request for preliminary ruling raised by the Sofia City Court in the context of confiscation proceedings under the 2018 Bulgarian Anti-corruption Law, a law which allows for the civil confiscation of assets prior to and irrespective of a criminal conviction. In the specific circumstances of the case, BP, a corporate bank officer, was subject to a criminal investigation for inciting the embezzlement of 105 million euros from the bank. Pending the criminal investigation, separate and independent civil proceedings under the Anti-Corruption Law were instituted before the referring court for the confiscation of assets that it appeared BP had obtained illegally.

In this context, the referring court sought a preliminary ruling of the Court of Justice by asking whether such form of civil and non-conviction-based confiscation was compatible with Directive 2014/42/EU establishing minimum rules on the freezing and confiscation of property in criminal matters (‘the Directive’) and partially replacing the pre-existing Framework Decision 2005/212/JHA (‘the 2005 Framework Decision’).

Under the Directive, confiscation presupposes that there is a final conviction for a criminal offence (Article 4(1) of the Directive), except for cases where such conviction was not possible due to the illness or the absconding of the accused (Article 4(2) of the Directive). The original proposal for the Directive submitted by the European Commission back in 2012 contained a provision on non-conviction based confiscation that was dropped during the negotiations between Council and European Parliament.

Against this background, should the Directive be interpreted as precluding confiscation in the absence of a final criminal conviction?

In answering the question, the Court of Justice made some preliminary considerations about the relevant EU legal framework.

In agreement with the Opinion of Advocate General Sharpston, the Court of Justice first observed that the Directive was not applicable ratione materiae to the case at stake, since its scope of application (Article 3) is limited to a list of offences which does not include embezzlement. This circumstance, however, did not make the preliminary question per se inadmissible.

The Court of Justice indeed referred to the 2005 Framework Decision as the EU legal instrument relevant for carrying out its interpretation of the question, noting that such an instrument had only been partly amended and repealed by the Directive and its untouched parts still remain in force as potentially applicable EU law (interesting considerations in this regard were made by Advocate General Sharpston in her Opinion). Differently from the Directive, the scope of application of the 2005 Framework Decision is not limited to specific offences but extends to the confiscation of instrumentalities and proceeds of criminal offences punishable by deprivation of liberty for more than one year. The offence of embezzlement under Bulgarian Criminal Code meets this threshold.

The Court of Justice then carried out an assessment of the 2005 Framework Decision in light of its objectives and its original pre-Lisbon legal basis (Article 31(1)(c) TUE) highlighting the minimum harmonisation ambitions of the instrument and its instrumental nature to the mutual recognition of confiscation decisions adopted in a criminal proceedings (to this effect, the Court stressed the link between the 2005 Framework Decision and Framework Decision 2006/783/JHA on mutual recognition of confiscation orders*).

This latter consideration led the Court of Justice, rather quickly, to exclude civil confiscation proceedings such as the one at stake before the referring court from the scope of application of the 2005 Framework Decision. Consequently, civil confiscation proceedings are not currently governed or precluded, by EU law.

From a broader perspective, the Court of Justice’s judgement is particularly relevant and to a certain extent striking.

In first place, the judgment is a reminder of the complexity and the difficulties of capturing the nature of confiscation, which remains a ‘many-sided concept’.** Even accepting that not all forms of confiscation can be classified as criminal punishment, it appears somehow striking that the Court did not consider carrying out an assessment of the substantive nature of the Bulgarian civil confiscation proceedings in light of its own developed Bonda criteria (a set of criteria which mirror the famous ECtHR Engel criteria and allow for the establishment of when a penalty or proceedings can be deemed ‘criminal in nature’ even beyond their formal qualification). Concerns about the respect of the principle of conferral and procedural autonomy may have played a role in the Court’s approach, but the absence of any consideration in this regard leaves a commentator reading it for the first time with a feeling of incompleteness.

In the second place, it has to be recalled that non-conviction based confiscation is receiving increasing attention at policy level and that several different forms of civil, in rem or non-conviction based confiscation already exist in the Member States.*** While these tools are deemed particularly effective against the most serious forms of criminality, they also require an adequate balance and regulation to remain compliant with fundamental rights (although not always in unambiguous terms, the case law of the European Court of Human Rights has dealt with non-conviction based confiscation on several occasions: see, for instance, the GIEM case).

In the present judgment, the Court of Justice ruled that EU Law does not preclude non-conviction based confiscation, but the question remains about whether and how EU law should regulate such measures in order to guarantee certain minimum procedural guarantees across the EU.

A reflection in this sense may be even more necessary in light of the new Regulation 2018/1805 on the mutual recognition of confiscation, which seems to extend the possibility of mutual recognition to certain forms of non-conviction based confiscation. The new Regulation (Article 2(1) and (2)), defines a ‘confiscation order’ as a penalty or a measure imposed by a court following ‘proceedings in relation to a criminal offence’ and not just in the context of criminal proceedings; such a terminological choice, according to some commentators,**** could include the mutual recognition of certain forms of non-conviction based confiscation, such as the Italian confisca di prevenzione.


*This Framework Decision shall be replaced by Regulation 2018/1805/EU as of 18 December 2020.

**For a thorough reflection see Panzavolta, Confiscation and the concept of punishment: can there be confiscation without conviction?, in Ligeti and Simonato (eds) Chasing Criminal Money. Challenges and Perspectives on Asset Recovery in the EU,  2017.

***See, for instance, the Analysis on non-conviction based confiscation measures in the European Union published by the European Commission in April 2019 (SWD (2019) 1050).

****Maugeri, Il Regolamento (UE) 2018/1805 per il reciproco riconoscimento dei provvedimenti di congelamento e di confisca: una pietra angolare per la cooperazione e l’efficienza in Diritto Penale Contemporaneo, 2019.


Angelo Marletta is a post-doctoral researcher in Criminal Law at KU Leuven. His research interests lie in European Criminal Law, with a particular focus on mutual recognition and the protection of the EU’s financial interests, and he is author of several publications in the field  (among the most recent, Ligeti and Marletta (eds.), Punitive Liability of Heads of Business in the EU: A Comparative Study, 2018).


Your privacy is important for us

We use cookies to improve the user experience. Please review privacy preferences.

I agree

Check our privacy policy and cookies policy.