Accountability Gaps in the Single Supervisory Mechanism (SSM) Framework

The first judgement of the General Court (EU) pertaining to the Single Supervision Mechanism (SSM) raised an issue concerning the source of the shared supervisory powers in the system of EU prudential supervision. In the General Court’s view, the SSM Regulation conferred on the European Central Bank (ECB) exclusive powers to carry out specific prudential supervisory tasks vis-à-vis all euro area credit institutions (significant and less significant). At the same time, the ECB’s national counterparts assist the former in implementing its supervisory tasks vis-à-vis less significant credit institutions. We argue that such a scheme requires a clear accountability relationship also between the ECB and the national competent authorities (NCAs), something that is not fully fledged in the current legal framework.


I. Introduction
Prudential supervision in the EU has drastically changed since the 2014 entry into force of Regulation EU/1024/2013 establishing the Single Supervisory Mechanism (SSM Regulation). 1 According to Article 4(1) of the SSM Regulation, the European Central Bank (ECB) assumed the role of the exclusive prudential supervisor over the euro area credit institutions. The ECB did not, however, fully replace national competent authorities (NCAs). NCAs assist the EU authority in discharging its supervisory tasks. 2 Indeed, banking supervision has become very much shared between the EU and national level authorities and staff. 3 The ECB has taken control over the 'significant credit institutions' (SIs), i.e. 120 groups representing approximately 1,200 supervised entities (out of 4,700 in total). 4 At the same time, NCAs oversee the 'less significant credit institutions' (LSIs). Moreover, concerning the supervision of SIs, the joint supervisory teams (JSTs) -entrusted with the dayto-day supervision of individual banks -are composed of EU and national staff members. 5 The system of banking supervision has thus truly become 'shared'. While (academic) research has given significant attention to the SSM, its operation and its accountability mechanisms, 6 little attention has been given to the exact interrelationship between the EU and national levels as regards the division of competencies. Where do which powers lie in this composite system of banking supervision? This is important for determining such issues as the system of accountability and of judicial protection. At what level does accountability need to be ensured? How? To what extent does the 'sharedness' of tasks challenge the organisation of controls and the present architecture of the EU in general? Accountability and judicial protection are essential for ensuring the rule of law. 7 They promote control over the exercise of public power and prevent arbitrary interferences with the rights and freedoms of private actors. The notion of accountability may come in various forms. Here we focus on political and judicial accountability, i.e., before representative and judicial institutions respectively.
Elsewhere, we have labelled the distribution of competences between the ECB and NCAs as hierarchical. 8 What skipped general attention so far, seems to be where the supervisory powers come from. At first glance, the question may seem obsolete. The SSM Regulation passed by the Council on the basis of Article 127(6) of the Treaty on the Functioning of the European Union (TFEU) introduces the SSM, outlining the powers of the ECB and of the NCAs. Therefore, the powers have been conferred from the national level(s) to the EU legislature, which subsequently divided the powers between the ECB and the NCAs. However, the issue has appeared to be much more complex, as has been discussed in the first judgement of the General Court pertaining to the operation of the SSM (Case T-122/15 Landeskreditbank Baden-Württemberg v ECB). 9 In the Court's view, the SSM Regulation entrusted all supervisory powers to the ECB (thus not to both the ECB and to the NCAs). That being said, national authorities play an assisting role, acting under the ECB's command. This postulation, which essentially refers to the source of supervisory powers and as a consequence to the relationship between the ECB and NCAs under the SSM framework, seems to suggest that we are now dealing with a 'reverse' delegation doctrine in the EU, i.e., the delegation of EU powers (supervision over LSIs) by an EU institution (the ECB) to the ECB's counterparts at the national level.
We argue that the Court's articulation carries implications for how such a delegation ought to be organised and accounted for. The current legislative framework of the SSM regulation does not, however, appear fit for purpose.
This article proceeds in the following way. First, a brief account of the new system of banking supervision in the EU, its origins and powers (section II) is given. Second, the authors discuss the abovementioned judgment of the General Court, by highlighting the facts, main arguments and main findings of the Court (section III). Third, the article zooms in to the part of the judgement where two scenarios of the source of the supervisory power are outlined and discusses the implications of the Court's choice for accountability in this respect (section IV). In section V, we conclude.

II. SSM: Origins, Legal Basis and Distribution of Competences According to the SSM Regulation
The idea of creating a pan-European watchdog for the supervision of euro area banks dates back at least to 2008. Following a number of crises which posed serious threats to the financial stability of the Euro area, the EU Commission mandated an expert group to examine the causes of the crisis and bring forward proposals for establishing an ' efficient, integrated and sustainable European system of supervision'. 10 The report (de Larosière report) called for radical reforms in the financial services sector, with a particular focus on supervision. 11 The potential role that the ECB could play in the attempt to centralise and coordinate the supervision of cross-border banks in the EU is mentioned several times throughout the de Larosière report. 12 The expert group, however, did not embrace such views, as -according to it -conferring micro-prudential task on the ECB would interfere inter alia with its monetary objective. 13 Despite the foregoing, in June 2012, the necessary political consensus was achieved and thus -since November 2014 -prudential supervision in the euro area is carried out under the aegis of the ECB. The legal instrument adopted by the Council is Regulation EU/1024/2013 conferring specific tasks on the ECB concerning policies relating to the prudential supervision of credit institutions (SSM Regulation). The legal basis of the SSM Regulation is Article 127(6) TFEU. While this legal basis is not undebatable, 14 opting for entrusting supervisory powers to an EU institution seemed advantageous, at least for two reasons. First, it would not provoke arguments revolving around the Meroni doctrine, which relates to the limits of delegating discretionary powers to agencies. 15 Second, setting up a centralised, pan-European system of banking supervision would -arguably -contribute to putting an end to the vicious circle of sovereign-bank doom loop. 16 When it comes to questions of hierarchy, the overarching principle in the SSM Regulation is that the ECB is exclusively responsible for the effective and consistent functioning of the SSM. 17 For example, NCAs must follow the instructions of the ECB, 18 which implies a hierarchy between the EU institution and its national counterparts.
As already mentioned (Section I), the SSM is premised upon a distinction between significant (SIs) and less significant (LSIs) credit institutions. The ECB is exclusively responsible for supervision of SIs as far as the tasks that are enlisted in the SSM Regulation. At the same time, according to the same legislative instrument, NCAs retain oversight over the supervision of LSIs. 19 Significance is primarily assessed on the basis of a bank's size, its importance for the economy of the EU or a participating Member State, the significance of its cross-border activities, and the total value of its assets. 20 According to Articles 70 and 71 of the SSM Framework Regulation, which lays down the framework for cooperation between the ECB and NCAs, 21 if particular circumstances so justify, a credit institution otherwise classifiable as a significant one, may be classified as a LSI, thereby falling under the supervision of the relevant NCA. Particular circumstances exist 'where there are specific and factual circumstances that make the classification of a supervised entity as significant inappropriate' 22 in view of the objectives of the SSM Regulation. However, the term "particular circumstances" is to be interpreted in a strict manner, 23 and the assessment thereof made on a case-by-case basis. 24 This implies that the ECB generally enjoys discretion in carrying out such an assessment.
The inverse case, that of the ECB assuming direct supervision of a LSI, is also possible. 25 Article 6(5) (b) elaborates on the circumstances under which the ECB can take over the supervision of an LSI: when this is necessary to ensure consistent application of high supervisory standards. This could be the case where financial assistance has been received from the EFSF or the ESM. 26 Indeed, how the ECB is to assess significance became a bone of contention in the Landeskreditbank Baden-Württemberg v ECB (L-Bank case), 27 where the General Court also touched upon the functioning of the SSM.

III. Case T-122/15 Landeskreditbank Baden-Württemberg v ECB: Key Issues
The L-Bank case is an important judgment to the extent that it clarifies the source and the scope of the supervisory powers that are vested in the ECB and the NCAs. Of significant importance are the General Court's articulations concerning the role of the opinion of the ECB's Administrative Board of Review in observing the obligation to state reasons. 28 Let us start with facts that led to the dispute. In June 2014, the ECB informed L-bank that on the basis of its size and in the light of Article 6(4) SSM Regulation, it had been placed under the direct supervision of the ECB, in other words, it had been classified as a SI. Subsequently, it invited the bank to submit its observations on its identification as a significant institution. In July 2014, L-bank submitted observations arguing inter alia the existence of 'particular circumstances' to be classified as a LSI, within the meaning of Articles 70 and 71 of the Framework Regulation.
Eventually, the ECB adopted a decision, which confirmed the classification of the L-bank as a SI. In October 2014, L-bank sought review of the ECB decision by the Administrative Board of Review (ABoR), i.e. the ECB's internal mechanism for review. The ABoR endorsed the ECB decision and as a result, in January 2015, the ECB repealed and replaced its decision dating to 2014, which ultimately meant that the L-bank remained for another year under the direct supervision of the ECB. L-bank contested this last decision before the CJEU, by bringing forward five pleas in law. It is worth mentioning that in arguing that all pleas should be rejected, the ECB was supported by the EU Commission.
By its first plea in law, 29 L-bank questioned the scope of the competences transferred to the ECB by the SSM Regulation in light of the principles of subsidiarity and proportionality. Specifically, the applicant argued (a) that the ECB interpreted incorrectly the conditions under which a bank's classification as significant is 'inappropriate' under Article 70(1) SSM Regulation; (b) that the ECB was wrong in classifying the bank as significant because it did not take into account the objectives and principles of the SSM Regulation; (c) that the ECB erred in law in its interpretation of the concept of 'particular circumstances'.
In answering the first plea, the General Court found it necessary to look into the scope of the competences transferred to the ECB by the SSM Regulation. It thus noted that a combined reading of Articles 4(1) and 6 of the SSM Regulation suggested that the creation of the SSM does not lead to a distribution of competences between the ECB and the NCAs for task therein. 30 (Article 4[1]). It rather provides for the 'exclusive competences delegated to the ECB to be implemented within a decentralised framework'. 31 To support this view, the General Court refers to some examples, such as to the fact that the ECB may decide to remove direct prudential supervision of a bank from the competence of an NCA. 32 In addition, the Court interpreted Article 70(1) SSMR through the prisms of subsidiarity and proportionality. As is well known, the principle of subsidiarity is relevant in areas in which the EU does not have exclusive competence. According to the General Court, by invoking the principle of subsidiarity, the L-bank essentially questioned the scope of the competences transferred to the ECB by the SSM Regulation. In other words, L-bank argued inter alia that the SSM Regulation should not be read as meaning that there has been a transfer of competence to the ECB for all the prudential tasks enshrined in Article 4(1) SSM Regulation and in relation to all credit institutions. The ECB, on the other hand, supported by the EU Commission, argued that it was granted exclusive competence to carry out all of the tasks contained in Article 4(1) SSM Regulation and that only the implementation of the tasks laid down in Article 4(1)(b) and (d) to (i) in relation to less significant credit institutions was delegated to NCAs under the supervision of the ECB. 33 The Court responded that the Council has delegated to the ECB exclusive competence in respect of all prudential tasks laid down in Article 4(1) SSM Regulation. Thus, according to the General Court, the prudential supervision carried out by NCAs under the SSM is to be seen as a mechanism of assistance to the ECB rather than the exercise of an autonomous competence. 34 The rationale for entrusting the prudential supervision of less significant banks to national authorities was to allow for ' decentralised implementation' of by the national authorities of the ECB's exclusive competence. 35 As to the principle of proportionality, L-bank argued that the implementation of an exclusive competence of the Union must thus be executed in a manner that leaves as much leeway as possible for the exercise of national competences. The General Court concluded that the ECB decision was not contrary to the proportionality principle because Article 70(1) of the SSM Framework Regulation refers to specific factual circumstances which make direct supervision by a national authority more fit for purpose with a view to attaining the objectives of the SSM Regulation. By its second plea, L-bank argued that the contested decision contained manifest errors of assessment. Specifically, it argued that direct supervision by the ECB was not necessary to ensure the safety and stability of financial markets, it was not necessary to safeguard the consistent application of supervisory standards and it was not necessary in view of the other objectives of the SSM Regulation. Finally, L-bank asserted that supervision by the German NCA would comply with the principles of the SSM Regulation, given that the bank is established under German law. 37 The General Court dismissed the second plea, on the grounds that the applicant failed to demonstrate that supervision by the German authorities could attain the objectives laid down by the SSM Regulation better than supervision by the ECB. 38 In addition, the General Court noted that L-bank failed to highlight any arrangement between it and the relevant German competent authorities that would render cooperation with these authorities easier than cooperation with the ECB. 39 By its third plea in law, the applicant asserted that the ECB infringed its obligation to state reasons. To back this plea, it brought forward three reasons. First, it argued that the ECB decision was inconsistent and self-contradictory, in the sense that it failed to explain which criterion was applied by the ECB. Second, that the statement of reasons accompanying the contested decision failed to explain why the absence of risk in the bank's activities was not a 'particular circumstance' and why supervision by the German NCA would not attain in a better way the objectives of the SSM Regulation. Finally, L-bank criticised the ECB for not explaining why certain arguments invoked by L-bank during the administrative review phase were not found to be relevant. To stress this point, the bank emphasised the discretion that the ECB enjoys in applying Article 70(1) of the SSM Framework Regulation; that the ECB enjoys such discretion, rendered a detailed explanation of how Article 70(1) SSM FR was applied all the more essential. 40 After discussing the relevant provisions establishing the ECB's obligation to state reasons found within the SSM legal framework, Union law and case law, 41 the General Court referred to the opinion of the Administrative Board of Review (ABoR). It should be noted that an ABoR opinion is not binding upon the ECB. 42 The General Court, however, -considered ABoR's opinion to be of relevance in this regard, as it -in essence -constitutes an extension of the ECB's decision. The General Court responded that if an ECB decision endorses the proposal set out in an ABoR opinion, the explanations contained in that opinion may also be taken into account in assessing whether the contested ECB decision contains a sufficient statement of reasons. 43 A combined reading of the ECB decision and the ABoR opinion demonstrated that the obligation to state reasons had not been infringed.
By its fourth plea, the applicant complained that the ECB misused its power by failing to exercise its discretion in the application of Article 70(1) SSM Framework Regulation. 44 Specifically, L-bank noted that Article 70(1), which lays down the concept of particular circumstances under which a SI can be classified as a LSI, does not contain an exhaustive list of reasons that the ECB may take into consideration. According to L-bank, the ECB was, therefore, incorrect in deciding that the arguments put forward by the supervised entity led to reasons not provided for in the SSM Regulation being taken into account. The General Court responded that the bank's argument put forward during the administrative procedure, namely that direct supervision by German authorities could attain the objectives of the SSM regulation because of the bank's low risk profile, was completely irrelevant for the application of Article 70(1) SSM Regulation and thus also rejected this plea. 45 As posited by the General Court 'the ECB cannot […] be criticised for having failed to exercise its discretion by rejecting at the outset an argument that is completely irrelevant'. 46 By its fifth and last plea, L-bank argued that the ECB infringed its obligation to examine and take into consideration all the relevant circumstances of the case, an obligation which is derived from the fundamental right to good administration. 47 The 'relevant circumstances' that -according to the applicant -were not taken into consideration were (a) that L-bank met no criterion enshrined in Article 6(4) SSM Regulation 37 ibid. para 102. 38  other than size and (b) previous supervision by German authorities had not displayed any shortcomings. Again, for reasons explained earlier in the judgment, the Court found these arguments to be irrelevant in view of the wording of Article 70(1) SSM Framework Regulation.

IV. The Source of Supervisory Power and the Implications Thereof for Accountability
The part of the judgement analysed in this article concerns the scope of the competences transferred to the ECB by the SSM Regulation. The issue may not seem significant, but the implications that it carries for accountability in the SSM framework are more substantial than one may think. At a meta level, this discussion raises a more general question concerning the EU governance architecture. This architecture is increasingly merged in terms of tasks and powers that both EU and national authorities perform, yet the systems of controls remain separated. This can lead to gaps in accountability. 48 In the L-Bank case, the applicant questioned 'that there has been a transfer of competence to the ECB with respect to all of the tasks referred to in Article 4(1) of the Basic Regulation and in respect of all entities'. 49 The L-bank distinguished between direct and indirect supervisory tasks, whereby the former were given to the ECB as they related to the supervision of SIs only, while the latter were given to the NCAs as far as they concerned LSIs. This position differed from that of the ECB's and of the EU Commission's, who took the view that the ECB is entrusted with carrying out supervisory tasks over all euro area banks, while NCAs were given certain implementation tasks enlisted in Article 4(1)(a) and (c) of the SSM Regulation. However, what these two views have in common is that they both see the SSM Regulation as the source of supervisory powers attributed to the ECB and to NCAs in the new SSM system (Figure 1. Scenario 1). The General Court concluded something else (Figure 2. Scenario 2).
In the General Court's view, the Council -by means of the SSM Regulation -has conferred an exclusive competence to the ECB to carry out specific prudential tasks over all euro area banks. 50 The SSM Regulation does not establish a distribution of competences between the ECB and the NCAs; it 'simply' introduces a decentralised framework for the implementation of the ECB's exclusive tasks. The General Court supported its view by, among other things, the existence of such powers of the ECB as the power to determine 'particular circumstances' when the ECB wishes to take over supervision of an LSI. As another example would be the power to issue regulations, guidelines or general instructions to NCAs. The General Court concluded, 'the existence of such prerogatives is indicative of the subordinate nature of the intervention by the national authorities in the performance of those tasks'. 51 All in all, the ECB is in charge and the NCAs assist the ECB under the latter's command.
The difference between the two scenarios is the source from where the NCAs get their SSM-related powers. This influences the relationship between the ECB and NCAs; the organisation of accountability relationships between them (internal accountability), and the accountability system for the SSM operation (external accountability).  Under scenario 1, the ECB and relevant national authorities derive their supervisory powers from the SSM Regulation, which implies a parallel division of competences (supervision for SIs and LSIs) with possible guidance from the ECB for the supervision of LSIs. Supervision, where necessary, upholds the stability of the whole system. In such a system, it seems to be necessary that the system of controls is organised similarly in a parallel way, at each level. At the EU level, the ECB is accountable before the European Parliament and the Council. 52 Additionally, the ECB is accountable before national parliaments. 53 NCAs are accountable at the national level before their national parliaments. 54 The ECB can give guidance to the NCAs on the implementation of their tasks under the SSM Regulation and NCAs can be requested to report back on the performance of their duties. Indeed, the system of controls under the current SSM Regulation corresponds to this description. 55 Additionally, the SSM Regulation provides for a specific mechanism of accountability of the ECB vis-à-vis national parliaments. 56 The General Court, however, favours scenario 2. This Scenario implies that the ECB is the exclusive recipient of all powers contained in the SSM Regulation, with NCAs assisting in carrying these out. Therefore, in the view of the General Court, the SSM Regulation delegates (emphasis added) exclusive competences to the ECB to be implemented in a decentralised system. 57 The General Court classifies the relationship between the ECB and the NCAs as 'hierarchical', in that the scope of powers of the NCAs, as well as the guidance on how they should carry out those powers, are determined by the ECB. A process of giving and taking back someone else's tasks and determining the scope or operation of said tasks is a delegation of tasks. However, in light of the traditional direction of the delegation of powers in the EU (from the national to the EU levels), the delegation line between the ECB and NCAs seems to be a 'reverse' delegation line. Given the logic of a delegation-accountability line, 58 the system of controls requires an accountability line between the ECB (' delegating authority') and the NCAs ('recipients of the delegated powers'). The full presence of that link under the current legal framework is, however, questionable.
To determine the link, we use one of the established definitions and assessment frameworks on accountability, that of Bovens. 59 Three stages are necessary to be present in an accountability relationship between an accountability actor (an NCA) and a forum (the ECB), i.e. an information, discussion and rectification stage: 'Without information, discussion is futile; having information without a possibility of discussion can prevent the rectification of mistakes; accountability without sanctions is incomplete as the 52 SSM Regulation (n 1) art 20(1). 53 ibid. art 21. 54 ibid. art 21(4). 55   presence of a possibility to sanction makes the difference between noncommittal provision of information and being held to account'. 60 The SSM legislative framework establishes the possibility for the information and discussion stages of accountability, but not for the rectification stage. According to Article 6 SSM Regulation, NCAs shall follow the instructions of the ECB and report on the performance of its duties, both on a regular and an ad hoc basis. Besides the possibility enshrined in Article 271(d) TFEU, according to which the Governing Council of the ECB may bring an NCA before the CJEU, should the latter fail to fulfill its obligations stemming from the Treaties and from the Statute of the European System of Central Banks and of the ECB, 61 the ECB has no real power to 'sanction' an NCA for poor performance or for not following its instruction. Furthermore, the relevance of Article 271(d) TFEU is questionable, as it was drafted in the context of monetary policy and long before the creation of the SSM. The only indirect 'punishment' vis-à-vis a poorly performing NCA is that of the ECB taking over the supervision of an LSI. Concerning the power of the ECB to instruct NCAs, it is unclear what the exact legal form of this power is. This can equally influence external controls, such as judicial review if the instruction is not a reviewable act. 62 A clear accountability link between the ECB and the NCAs is essential for the proper exercise of the ECB's external accountability before the European Parliament (EP) or the Council, and for the correct functioning of the SSM system where the ECB is an exclusive recipient of all the powers. Under the current regime, the ECB is first requested to render account for the operation of a system that it may not be able to control fully without an instrument of directing and punishing individual NCAs. 63 Secondly, it can point fingers towards an NCA when rendering account at the EU level, thus shifting the blame for a failure. The fact that NCAs render account for their assistance to the ECB before the national parliaments cannot fully compensate for the weak link between the ECB and the NCA. Here too the possibility to point fingers to the ECB is a possible gap. Finally, given the possibility of potential divergence between EU/ECB and national (political) interests, accountability before the national parliament may not necessarily equate accountability by the ECB in terms 'for what', 'how' and 'how heavy' it exercised its powers. Now, how to remedy the existing situation to match the necessities of scenario 2 and to ensure accountability? The ECB and NCAs should have a strong accountability relationship. The ECB should particularly be able to delineate the tasks of the NCAs. It should also be able to follow the execution of these tasks; this is already provided for by the legal framework to a certain extent, via the reporting obligations on the part of the NCAs. 64 Lastly, the ECB should be able to sanction them for poor performance. This is currently lacking, but in this way, the ECB will be able to render account for the operation of SSM before the EP and the Council.
First, it would be helpful to determine the legal form of ECB instructions, because this predetermines the necessity and possibility to use one or more accountability channels. Will it become available to the public/private actors mentioned in that instruction? This would facilitate judicial accountability, as the private parties would arguably be able to challenge it before a court. Will the instruction affect a large number of credit institutions or will it be addressed to an individual bank? 65 The answer to this question determines whether an affected bank would have standing before a court. Does the NCA to which an instruction is addressed enjoy discretion in executing it? The answer to this question determines which court, EU or national, will be competent to review the instruction ex ante. In addition, the adoption of such instructions may require the establishment of necessary (administrative) procedures within the ECB. If instructions are used only between the ECB and NCAs, the political accountability of the ECB for such instructions before the EP and the Council (and perhaps even before the relevant national parliament) should be enhanced. Currently, the legislative framework is not explicit on these issues, which may lead to accountability gaps.
Second, the rectification mechanism for not following the instruction should be available for the ECB vis-à-vis an NCA which does not follow it. 66 Generally, several options could be available in such cases at the national level, namely internal administrative, political or legal accountability. Internally, the head executive could remove a top official of an NCA from office. Politically, the national parliament could reduce the budget and/or functions of the NCA. 67 These options seem to be less feasible in the composite system of governance where the ECB does not appoint the NCAs' top level officials and the accountability before the national parliament is as mentioned before not necessarily a fully-fledged substitute given the EU vs. national interest background. As a soft internal administrative sanction, the ECB could potentially appoint an internal watchdog similar to those used in the US. 68 Inspector Generals are placed by Congress in all major independent agencies and departments of ministries to supervise, coordinate policies and keep Congress informed about possible problems. 69 In any case, a final option is that of legal accountability. One could think of using the possibility contained in Article 271(d) TFEU, which resembles the infringement procedure of the EU Commission against Member States (Art. 258-260 TFEU), the difference being that by virtue of Article 271(d), the ECB can bring an NCA before the CJEU. More importantly, the taking over of supervision of an LSI may be seen as a sanction against an NCA/MS. However, it is worth considering that this also affects the private actor and the operational ability of the ECB to supervise all credit institutions in the first place.

V. Conclusion
According to the General Court, the SSM Regulation entrusted the ECB with exclusive powers of prudential supervision over the euro area credit institutions. The NCAs assist the ECB in discharging its tasks. The General Court has classified this relationship as 'hierarchical'. We have argued that the relationship between the ECB and the NCAs involves a 'reverse' delegation line; such a link requires an accountability relationship between the ECB and the NCAs, which is not fully-fledged at this moment. The legal form of the instructions needs to be clarified, as it carries implications for the types of accountability that could be available for different actors (private parties, political forums). In addition, the ECB needs to have a rectification mechanism against NCAs which do not follow its instructions. At this moment, it lacks a tool to sanction an NCA for poor performance or non-compliance and the accountability of an NCA before its national parliament may not necessarily compensate for the 'weak' link between the ECB and the NCA.
This discussion also raises a more general concern about the suitability of the EU's ' constitutional' architecture to the realities in which it operates. The SSM framework is only one example of a rapidly developing 'mixed/shared/integrated administration' of rule-making, supervision and enforcement. The trend increasingly merges the functions of, roughly speaking, two separated legal orders -that of the EU and that of EU Member States. This has been developing without necessarily being accompanied by modifications to the Treaties. 70 Importantly, the merging of tasks occurs whilst the system of controls between two legal orders remains separated. This creates incomplete or weak accountability systems and should deserve our attention in research and practice.