SOMMARIO[*]: Le implicazioni pratiche del principio generale di proporzionalità per le misure di intervento precoce sono significative e sia i tribunali che i legislatori si stanno adoperando per chiarirle. Questo breve saggio offre una visione informata che guarda indietro, nel presente e in avanti, ai diversi livelli normativo, di supervisione e giudiziari, tenendo conto del significato multidimensionale del principio di proporzionalità.
ABSTRACT: The practical implications of the general principle of proportionality for early intervention measures are significant, and courts and policy makers are currently endeavouring to clarify them. This brief essay offers an informative view looking backward, into the present and forward, at the distinct regulatory, administrative and judicial levels, factoring in the multi-dimensional significance of proportionality.
1. A premise
I will briefly discuss this very topical issue (i) looking backward, into the present and forward, (ii) at the distinct regulatory, administrative and judicial levels, (iii) factoring in the multi-dimensional significance of the super-principle of proportionality. I will do it in the hope to pour new wine into an old bottle, without much, though, in terms of academic argumentation to avoid that the bottle may burst in the end. For a proper academic discourse I refer to the informative (and still recent) works of David Ramos Muñoz (‘Twilight troubles’. Early Intervention, Resolution, Preparation and Triggers, and the Urgent Need to Reform Them, in Eur. Bus. Org. L. R., (2025) 26:133-155) and of Irene Mecatti (Early intervention measures and shareholders rights, EBI Working Paper Series, 2025 – no 192, of 23 July 2025). My quite modest attempt here is thus to offer what I dare to see as an informed view on Corneli, as the ongoing landmark case on this issue, and to briefly flag and earmark what I still see as problematic in the wake of the CMDI political compromise of end June 2025, judging from experience. With a final disclaimer and cautionary note: I have assisted as external lawyer the ECB before the Grand Chamber of the Court of Justice in Joined Cases C-777/22 P and C-778/22 P, Corneli. Though I am currently out of the case (the ECB was represented only by its agents before the General Court in the original Case T-502/19, that resulted in the judgment of 12 October 2022 ECLI:EU:T:2022:627 and is consistently represented only by its agents also in the current Case T-502/19 RENV, which has been resumed after the judgment of the Grand Chamber of 15 July 2025, ECLI:EU:C:2025:580), my view on this is also a reflection of this experience.
2. Looking backward
Looking backward, prompt corrective actions are embedded into the US banking regulation since the Federal Deposit Insurance Corporation Improvement Act of 1991. In turn, temporary administration has a long-lasting tradition in the Italian legal framework and has clearly informed the EU adoption of it by the BRRD. Since the global financial crisis the international regulatory framework has been significantly strengthened and there has been a common recognition, as noted by the Basel Committee on Banking Supervision in its March 2018 “Frameworks for early supervisory intervention” (at p. 4), that “for supervisors to operate effectively, identification and intervention at an early stage are critical to prevent problems from escalating or becoming acute” and that therefore (at p. 5) “early supervisory intervention is defined as supervisors taking actions at an early stage to address unsafe and unsound practices or activities that could pose risks to banks or the banking system. These early supervisory actions can range supervisory measures that encompass moral suasion to more corrective actions, which are triggered when banks are deemed to be in danger of failing”. Our focus here is on early intervention measures, and thus on the powers granted to supervisory authorities since 2014 by BRRD (art. 27-29) and SRMR (art. 16) to help timely address banks in bad shape yet not likely to fail, to seek to restore the situation to normality. In the European acquis, those measures sit in a continuum with supervisory measures under Article 102 ff. CRD (but also 91 and 104 CRD), showing – up to the incoming adoption of the CMDI – undesirable overlaps and ambiguities in their mutual relation as to triggers, framework for the escalation, and content. This has been admirably mapped and discussed already back in 2021 in the EBA Report on the application of early intervention measures in the EU in accordance with articles 27-29 of the BRRD (EBA/REP/2021/12): a piece of work that has well informed the CMDI regulatory process and namely the Commission’s proposal.
Crucial to our purpose is the question of the sequence in applying the EIMs under the principle of proportionality.
Looking backward, the EBA report, at § 75 ff., notes that the BRRD foresees a certain hierarchy in the application of EIMs stipulated in Articles 27, 28 and 29 BRRD, respectively. Pursuant to Article 28 BRRD, the competent authority may require the removal of the senior management or the management body of an institution if there is a significant deterioration of the institution’s financial situation or if there are serious infringements of certain legal provisions or serious administrative irregularities and if ‘measures taken pursuant to Article 27 BRRD are not sufficient to reverse the deterioration’. Correspondingly, a temporary administrator may be appointed pursuant to Article 29 BRRD if the replacement of the senior management or management body pursuant to Article 28 BRRD is deemed insufficient to remedy the situation. The EBA further recalls that there were some uncertainties related to the sequence of applying EIMs listed in Articles 27-29 of the BRRD, in particular whether it is necessary to actually apply the less intrusive measures to determine that they are insufficient to reverse the deterioration before more intrusive ones can be implemented. In order to remove this uncertainty, a Q&A was submitted to the EBA asking for clarification on the sequence of applying EIMs listed in Article 27 and 28 BRRD. The interpretation provided in the EBA Single Rulebook Q&A (Question ID 2015_2018) points out that it may not be necessary or possible to actually take the EIMs established in Article 27(1) BRRD before taking those in Article 28 BRRD. The Report adds, however, that despite this interpretation, the EBA survey revealed that the sequencing of the application of EIMs is problematic in some jurisdictions and does not correspond to practical needs in crisis situations. The sequencing concerns made the EIMs pursuant Article 28-29 EIMs difficult to implement in most cases, since their application process was very time-consuming and did not cover the risk of further deterioration the institution’s situation. Therefore, the EBA, in order to eliminate challenges related to the currently envisaged sequence of applying EIMs in accordance with Articles 27-29 BRRD, recommended to introduce one of the following solutions: Option 2.1: To merge all BRRD EIMs into one set of measures in Article 27 BRRD (eliminating any additional conditions for applying measures listed in Articles 27-29 BRRD in sequence). Measures currently listed in Articles 28 and 29 BRRD should be applicable at the same time as or instead of measures currently listed in Article 27 BRRD; Option 2.2: To merge EIMs from Article 28 and 29 BRRD into one set of EIMs that would remain in the BRRD while eliminating the sequencing of applying them. By contrast, Article 27(1) measures would be re-classified as supervisory powers. Option 2.3: To merge all EIMs (Articles 27, 28 and 29 BRRD) into one expanded set of supervisory powers included in the CRD/SSMR, eliminating the sequencing and additional conditions for applying measures currently listed in Articles 28-29 BRRD. The EBA further notes that the banking associations opposed the elimination of the existing sequencing of EIMs listed in Articles 27, 28 and 29 BRRD. They argued that the application of EIMs should follow a clear proportionality logic. Moreover, they believed that giving more discretion to supervisors in that matter would reduce legal certainty. This was also in line with the ECB position expressed in a letter of the Chair of the Supervisory Board to the Members of the European Parliament Lamberts and Urtasun of 24 January 2018, where the ECB pointed out that “from a proportionality perspective, it can be argued that early intervention measures are more intrusive than supervisory measures due to their role in crisis prevention and management, including the increased involvement of the resolution authority. For this reason, in situations where the competent authority needs to choose between takin a certain action as either an early intervention measure of as a supervisory measure, the principle of proportionality suggests that it should adopt a supervisory measure as the less intrusive option, provided that both types of measures are suitable to achieve the intended aim.” In other words, there is no doubt that the existence of a clear escalation ladder of supervisory action is a necessary application of the principle of proportionality in this domain; however, this does not imply (contrary to Irene’s belief, at page 7 of her recent paper, where she surmises that “collective removal powers can only be used after measures under Article 27 BRRD have been taken”: emphasis added) that more intrusive measures are available only after that less intrusive have been adopted and tested in practice (and have proved insufficient), because what really matters, in the logic of crisis prevention that permeates EIMs, is not the reality check but the duly substantiated assessment of the supervisor of the suitability or unsuitability of the less intrusive measures to achieve the intended aim of recovery without the necessity of a market test of such suitability. In the real world, banks’ crises may rapidly precipitate, and imposing to the supervisory authority to sequentially test in practice all early intervention measures before adopting the more intrusive ones would be tantamount as depriving of effectiveness the most powerful EIMs. This is a point that in the literature seems to be acknowledged already back in 2018 e.g. by Georgios Psaroudakis, Proportionality in the BRRD: Planning, Resolvability, Early Intervention, Beitraege zum Transnationalen Wirtschaftrecht, Heft 159, August 2018, at page 16, where he rightly notes that the express reference to proportionality in Article 28, where it presuppose that measures under Article 27 are not adequate, “has some procedural importance as it seems to impose on the competent authority a particular burden to justify the necessity of its decisions, i.e. the necessity of the chosen measure. It is a different matter, however, that the public authority retains some discretion in determining the proportionate response to the circumstances”. This is also echoed, in the Italian case-law, in the landmark judgment of the Upper Administrative Court (Consiglio di Stato) of 19 July 2022, removal at Credito di Romagna (at § 10.1.: “è evidente il carattere spiccatamente discrezionale del provvedimento in esame, volendosi con ciò sottolineare come l’autorità di vigilanza, una volta accertata l’esistenza dei presupposti oggettivi enunciati dall’art. 69-octiesdecies, co. 1, lett. B) TUB, sia chiamata a stabilire, sulla base di una valutazione di opportunità, se disporre la rimozione collettiva degli organi, ovvero se, nello specifico caso, siano prospettabili soluzioni meno gravose (o, se in ipotesi sia invece necessaria una misura ancora più incisiva). This conclusion should also guide, to my mind, in considering more restrictive national transposing laws, or national supervisory practices, which, as surveyed by the EBA (at § 78 of the Report), may have so far required the actual sequencing of the application of the EIMs in some jurisdictions: this may indeed run counter the EU general principles of effectiveness and proportionality (in its vertical dimension for the proper construction of the internal market: see on this the very insightful remarks of Chiara Zilioli, Proportionality as the organizing principle of European banking regulation, paper available on ssrn and based on the speech delivered on 13 February 2017 in Athens).
3. Looking into the present
Looking at this into the present, two aspects stand out to my knowledge.
a) On one hand, the CMDI compromise recently reached, which should pave the way for the inclusion of EIMs into the SRMR, and thus remedy, in the Banking Union, the undesirable situation that so far the ECB lacked a legal basis for the adoption of such measures in directly applicable Union law (something that eventually led, in Corneli, to the loaded question of the interpretation and application in conformity of national transposing law), offers clear guidance (to my mind, also working as an authentic interpretation of the existing regime, in light with the EBA Q&A). First, Article 13, in the compromise text, expressly states, in accordance with the Commission’s proposal and dropping an amendment of the European Parliament which would have made the adoption of the EIMs conditional upon “a significant deterioration of conditions, or new adverse circumstances or new information”, that the ECB may determine that the condition that remedial actions other than early intervention measures are insufficient to address the problems is met “without having previously taken other remedial actions”, including the exercise of the powers referred to in Article 104 CRD. This would be complemented by new Article 13(3), whereby “the ECB shall choose the appropriate early intervention measures based on what is proportionate to the objectives pursued, having regard o the seriousness of the infringement or likely infringement and the speed of deterioration in the financial situation of the entity, among other relevant information”. Second, new Article 13b applies the proportionality principle also to guide the ECB in the choice between a temporary administrator that either a) temporarily replaces the management body of the entity or b) works temporarily with the management body of the entity. This is, once again, in line with Italian experience (as also tested, precisely vis-à-vis the requirements of proportionality and adequacy, in the Credito di Romagna case).
b) On the other hand, the Corneli case sheds light (in addition to its crucial findings on effectiveness of Union law, interpretation in conformity and nature of question of law of the related dispute on appeal) also on the application of the proportionality principle in the EIMs’ sequencing. To be true, the plea on the proportionality of the measure of temporary administration adopted by the ECB on 2 January 2019 is still currently pending before the General Court after the annulment with renvoi of its previous judgement of 2022. It is indeed the first plea in law of the original application for annulment, that was not examined by the General Court, which upheld the fourth plea on the legal basis of the measure. However, the Grand Chamber, at paragraphs 148-149 of its judgment of 15 July 2025 has clearly stated that:
148 (…) as regards Ms Corneli’s argument alleging, in essence, the need to make provision, in line with the principle of proportionality, for a ‘scale’ of intervention measures available to the competent authority in the management of a banking institution, it must be held that the system of intervention measures provided for in Articles 27 to 29 of Directive 2014/59 is in line with that principle.
149 As regards, specifically, the measure of temporary administration provided for in Article 29(1) of that directive, it is apparent from that provision that that measure may be adopted only after the less restrictive measure provided for in Article 28 of that directive, namely the replacement of the senior management or management body of the banking institution concerned, has been deemed insufficient in the light of the situation of that institution
Thus, it is now settled law that the proportionality principle does not require, in this context, that less intrusive supervisory or early intervention measures are adopted before temporary administration is adopted, provided that the supervisory authority has properly assessed and deemed that the other measures would have been insufficient in light of the situation of that institution. This is fully in line with past precedents in other fields, where European courts already held that the application of the proportionality principle, where several measures are available, whilst it requires an assessment in order to adopt the less restrictive measure which is sufficient to tackle the situation and is proportionate to achieve the intended aim (judgment of 18 September 2017, Joined Cases T-107/15 and T-347/15, Uganda Commercial Impex, ECLI:EU:T:2017:628, paragraph 125), does not require that all less intrusive measures are tested in practice (and thus adopted) before a more intrusive one becomes possible (judgement of 11 July 2007, T-170/06, Alrosa, ECLI:EU:T:2007:220, at paragraph 99).
Ultimately, it seems that, when interpreting these matters, we often look at the text, and then move to the safeguards, without looking at the context, and thus we fail to see the forest for the trees. Yes, the EIMs are more intrusive than supervisory measures, and yes, some EIMs are more intrusive than others. But EIMs must, first and foremost, be effective. Sitting uncomfortably between supervisory measures and resolution, either EIMs are available enough, and effective enough, or they will be squeezed out of existence. Thus, whereas proportionality is a principle that informs the whole crisis management framework as a balancing principle, one must balance something in the first place. This is effectiveness, and while effectiveness also permeates the whole framework, it is, contextually speaking, most important in the case of EIMs, precisely because there are other measures sitting on both ends of the spectrum.
4. Looking into the future
Looking into the future, it remains to be seen how the General Court shall assess, in its judicial review upon renvoi, the factual elements duly factored in in the ECB decision of 1 January 2019. On this, contrary to Irene’s claim, at page 19, that “the ECB could have resorted to a less restrictive instrument, specifically collective removal under Article 69-viciessemel, paragraph 1 TUB or, given the more serious conditions of Article 70 TUB, proceed to the appointment of temporary side-by-side commissioners”, I draw attention on to aspects, one factual and the other legal.
a) Factually, it is important to note that at the time of adoption by the ECB of the temporary administration decision the board of directors of Banca Carige was more than halved due to resignations of the majority of its members, with four of such resignations (including those of the Chair, Modiano, and CEO, Innocenzi, then appointed as two of the three temporary administrators) taking effect on 2 January 2019. Conversely, Article 18 of the articles of association of Carige set out a simul stabunt, simul cadent clause, and further specified that the the board remaining in office up to the appointment of the new board was constrained to the ordinary course of business. It seems evident that, in a rapidly deteriorating financial situation, the alternative EIM of the removal of the remaining members of the board would have been simply pointless (because the board would have fallen on the simul stabunt on the 2 January 2019 and was then bound to act temporarily in prorogatio until the appointment of the new board) and the appointment of a temporary administrator in addition to the board under Article 75-bis TUB would not have solved the issue of the constrained ability of the board in prorogatio to deal with matters outside the scope of the ordinary course of business; and if it had, assuming that the ECB could grant all such powers to the temporary administrator working temporarily with the remaining board members, this would have de facto conferred all relevant powers to the temporary administrator as it also happens with temporary administration replacing the board, but with much less clarity and certainty. It should also be considered, when factoring in all relevant attributes of temporary administration vis-à-vis other EIMs, including the removal, that only temporary administration, under Italian law, may come along with a suspension of payment under Article 74 TUB. A measure which may prove essential as part of the available toolkit to rapidly plan, prepare and implement recovery and prevent further liquidity outflows or payments in violation of the par condicio in the twilight zone.
b) Legally, the standard of review of European courts in checking compliance of administrative actions with the principle of proportionality, needs to acknowledge that, as I noted already elsewhere in the past, this is an area pervaded by the analysis and assessment of future scenarios. To ensure financial stability, authorities engage in supervisory, preparatory and preventative actions largely based on forward looking, speculative assumptions and in the EIMs domain, like in the resolution context, those assumptions are (remote, yet still possible) worst case scenarios. Upon a context filled with uncertainty and known unknowns, authorities needs to act when the risk of a ‘false negative’ clearly outweighs the risk of a ‘false positive’; or, to put it simply, where it is better safe than sorry. Therefore, whether it is stated explicitly or not, early intervention measures require combining more ‘familiar’ assessments, e.g., the conclusion that a certain event will occur in the future in a binary probabilistic assessment, with assessments that require making assumptions beyond a single probability distribution, to minimize the harmful consequences in each scenario e.g., “maximin with multiple priors”, or assessments that require weighing probabilities, but factoring in black swans and fat tail events. Given the uncertainty and the economic stakes, no surprise that these assessments become policy and legal battlegrounds. In this context, a key question is what evidential burden the supervisory authority has to bear in order to support complex technical assessments based on alternative options on hypothetical future scenarios. In antitrust, where fundamental rights concerns trump over any other considerations, there has traditionally been a call to treat ‘false positives’ (e.g. erroneous antitrust convictions and over-deterrence) as costlier than ‘false negatives’ (i.e. erroneous acquittals and under-deterrence), and ask for a higher evidentiary burden for those alleging an antitrust violation, through a ‘preponderance of evidence’ (in American terms) or ‘balance of probability’ in European terms. This cannot be extrapolated to supervision (or resolution) cases, where the financial stability implications of false negative are potentially catastrophic and financial stability concerns trump over any other considerations. Economic science cannot (at least cannot always) offer conclusive answers or full evidence of the shape and implications of potential future financial stability shocks, which may be used as input for resolution decisions. Thus, reasonableness and the ‘plausibility’ test are so far used by European courts to review supervisory and resolution measures. This is also in line, to my mind, with the proportionality check made by the Italian Upper Administrative Court (Consiglio di Stato) in the Credito di Romagna case, where on one hand the court rightly followed an hard look approach (also) on the factual premises of the supervisory assessment and appointed to that purpose, for the first time in this domain, an expert witness, but was conversely adamant in holding that courts cannot second guess nor replace with their own the complex technical assessment conferred upon the competent authority.
This approach places the justificatory burden of the resort to specific measures on the authority, but without excessively predetermining what the authority must justify. If experience teaches us something is that rapidly evolving situations, by definition, do not go according to plan. If we consider that the principle of proportionality, in the context of EIMs, implies primarily the need to use “less intrusive measures” first, we are not only hindering the effectiveness of EIMs (as argued before). We are also pretending to know too much about the challenges that the situation will present, and the scenarios and the choices that the authority will be confronted with. We want the authority to reason its way through the measures, but we also want the authority to reason through the real problems, not a bureaucratic list. Sadly, this is what will happen if we want to predetermine too much what proportionality means in advance.
A balanced review of proportionality in its application to supervisory measures sits well, to my mind, also with respect to the nature of such measures. Whilst it is certainly true that those measures may in the end affect indirectly individual fundamental rights of shareholders and creditors, it is too often neglected the fact that those measures are aimed at preserving, with the viability of the entity and financial stability, also those individual rights better than in the counterfactual of failure. This has important legal implications. Contrary to the (still to some extent dubitative) conclusions reached by AG Kokott in Corneli in her Opinion of 21 November 2024 (ECLI:EU:C:2024:973), this is a situation not different from administrative disputes which concern ‘triangular situations’. In this regard, European courts have acknowledged, by way of exception, that a third party must tolerate certain State measures on the basis of directly applicable provisions of a directive on which an individual relies vis-à-vis the Member State. These include, for example, an environmental impact assessment which can result in the revocation of a consent benefiting the third party or the loss of some other (legal and financial) benefit to the third party. This is merely an adverse repercussion of the implementation of such provisions on the legal position of third parties. In such a situation, proportionality requires also to weight the factual repercussions of EIMs against the alternative scenario (or counterfactual) of a likely failure, that would, in the end, affect more severely the very same individual rights.
Let me now conclude, by looking into the future, at what it may still be missing in the EU acquis on EIMs after the political compromise of the CMDI. Much has been already said, quite rightly, by David in his paper of 2025, and it suffices to refer to it. Yet, to my mind, now that the Unidroit guide on bank liquidation has been finally adopted, the co-legislators may find in the Guide ready-to-use solutions to address most if not all of the shortcomings identified by David. However, what still remains in a surprising cone of shade and is not remedied by the Unidroit guide (whose scope is by design more limited) is the interplay between EIMs, debt restructuring of the bank in bad shape via scheme of arrangements and restructuring plans agreed with the main creditors, the possible use of write down and conversion tool outside resolution as already envisaged by the BRRD, and a potential stay of payments along the lines of what is permitted by Article 74 TUB (such long stay being quite an exception in the international context, though). This would be an important area to look at, at European level, but also at national level, to clarify for instance the current uncertain application to banks’ crisis management of the so called “composizione negoziata della crisi” (negotiated crisis solution), which is not an insolvency procedure to the effect of Article 80, paragraph 6 TUB (which for banks has opted out the applicability of alternative general insolvency procedures), and may therefore also come into play for banks, bringing with it the possibility of court’s protective measures (and their long and protected stay of payment) as set out in the Insolvency code.
[*] International Colloquium on “Proportionating rules on bank crisis prevention and management to the case of retail banks: an analysis on the European and national legal framework, Venezia, Cà Foscari University, 11-12
