1. 268Introduction

The relationship between banking, competition and public intervention during crises is complex and tense.^[1] States do not want to let a bank fail for political considerations and as well as to preserve financial stability and market confidence. In turn, banks may take advantage of this expectation, engaging in excessive risk-taking and relying on implicit State guarantees. This interaction overall decreases competition in the banking market, reducing social welfare.

This tension reached its peak during the Global Financial Crisis that started in 2007. In Europe, this led to a series of massive bailouts throughout the whole Eurozone crisis, until 2012. To address this situation, the EU completely changed its policy, issuing a Banking Communication on State Aid in 2013 and establishing a new resolution framework for ailing banks, within the wider Banking Union project.^[2]

269Under the current regime, no public aid can be granted before a substantial private sector involvement – the so called ‘burden-sharing’ – which aims at reducing taxpayers’ exposure and ex-ante moral hazard from bankers. Moreover, any public aid to banks implies that the bank is ‘failing or likely to fail’ and should be resolved or liquidated accordingly.^[3]

This strict ‘no bailout’ policy should eradicate the problem of public ownership in banks, minimizing the distortions to the competitiveness of the banking market that comes with it.^[4] However, one decade later, this goal seems far to be reached, despite substantial steps forward. The recent banking turmoil of 2023 highlights that elements of flexibility in the use of public funds are perhaps unavoidable to limit financial stability concerns despite the distortions they bring about. Through these lenses, we can understand the massive emergency liquidity assistance provided by the Swiss National Bank to UBS and backed by the Swiss State to facilitate the acquisition of Credit Suisse.^[5] A similar argument goes for the triggering of the ‘systemic risk’ exception used by the US Congress to limit the run on uninsured deposits, allowing all banks to post collateral at par value for one year – the Bank Term Funding Program (BTPF) – resulting de facto in a massive bailout whose fiscal consequences are, to date, unclear.

The situation is further complicated by the fact that, to date there are still several distorted incentive mechanisms that create inefficiencies in the regulatory system. In particular, on the one hand, ample evidence shows that incentives for bankers are still leading toward the immediate recognition of the gains and the postponement of losses.^[6] Therefore, interventions in a distressed bank happen too late. In this regard, the recent cases of Credit Suisse and Silicon Valley 270Bank are paradigmatic cases of late intervention and distorted incentives.^[7] On the other hand, it is broadly acknowledged that private actors do not have incentives to take early action in the crisis.^[8] Moreover, the banking market in many countries, including the Euro Area countries, is illiquid, so that private solutions through mergers are often a non-viable solution.

Departing from this unsatisfactory state of facts, this article analyses the EU ‘precautionary recapitalization’ tool. This tool, holding a series of conditions, allows solvent but undercapitalized banks with no potential private buyer to apply for State aid, including in the form of capital injections. While this tool provides flexibility in handling crises, it also comes with costs. Some are straightforward, and namely, the direct fiscal cost of the aid and the increased moral hazard incentives for undercapitalized banks. However, there is a further key cost that has been overlooked by the academic debate and that represents the core of our analysis: the ex-post costs of having a bank in public ownership without being able to divest from it. In fact, while EU law stresses the ‘temporary nature’ of the precautionary recapitalization, first implementations highlight that such a temporary clause is difficult, if not impossible, to enforce, in line with the international evidence on the persistence of public ownership in banking and the complexity of divestment.^[9]

Having regard to this aspect, this paper assesses if and how – under the current regulatory framework – it is possible to enforce precautionary recapitalization’s temporariness when a market solution does not seem possible. We contend that the framework’s inability to ensure a prompt divestment poses a serious threat to the credibility of the whole resolution system and contradicts the principles governing the post-Global Financial Crisis regulatory system, imposing additional costs to taxpayers (a precautionary recapitalization which is, 271in fact, a bailout, usually requires additional public support) and reproposing moral hazard problems.^[10]

Therefore, this article aims at introducing new elements in the theoretical framework to analyze legal institutions dealing with State capital injection, applying it specifically to the EU precautionary recapitalization and its ‘temporary’ clause. The article disentangles the key junctures of such instrument, isolating the limitations that prevent the enforcement of its temporary nature.

This is relevant both from a theoretical and a policy perspective. From a theoretical standpoint, it provides a broader and better contextualized approach to State intervention in banking and, in particular, to precautionary recapitalization in Europe. To the best of our knowledge, this is the first article that clearly identifies and isolates the inability of ensuring temporariness as the key problem with the effectiveness of precautionary recapitalization. From a policy perspective, it also provides guidance on how to reform the instrument, informing the discussion over the CMDI whose first drafts appears utterly unsatisfactory from this standpoint.^[11]

The remainder of the article unfolds as follows. Section 2 sets down the analytical framework. Section 3 introduces the key legal regime for precautionary recapitalization. Section 4 analyses the cases of precautionary recapitalization in Europe. Section 5 disentangles the key junctures making precautionary recapitalization ex-post ineffective. Section 6 analyses the CMDI framework against the findings of the analysis. Section 7 concludes.

2. 272Ex-ante incentives and ex-post outcomes: public intervention in banking revisited

In the aftermath of the 2008 Global Financial Crisis, banking legislation worldwide has been reformed under the stimulus of international standard setters as the Basel Committee for Banking Supervision (BCBC) and the then-newly established Financial Stability Board (FSB). Bank resolution was at the core of this wave of reforms, with the aim to “make feasible the resolution of financial institutions without severe systemic disruption and without exposing taxpayers to loss while protecting vital economic functions through mechanisms which make it possible for shareholders and unsecured and uninsured creditors to absorb losses”.^[12]

This quote highlights all the key characteristics of the post-financial crisis consensus model to treat bank distress including the role of State intervention.^[13] This approach can be summarized in three key policy priorities. First, protecting the fair competition in the banking market, minimizing moral hazard of material risk-takers by limiting public intervention to the largest extent possible and, in any case, making sure that it does not excessively impact on ex-ante incentives of bankers; second, protecting taxpayers; third, protecting insured depositors and financial stability.

This approach is sensible and, despite many limitations in the implementation phase, represents a remarkable step forward in our understanding of banking and its regulation. However, this model completely overlooked the next stage, after the resolution of the distress took place, especially when public resources were involved. In other words, under the current framework it seems to be plainly assumed that the resolved bank is either liquidated or returned to the market. However, this is not the case and there is evidence that banks that received public aid in the form of capital injection tend to remain in public hands for a very long period of time.^[14] This severely limits competition, exposes taxpayers to further losses and lowers systemic resilience. In this regard, current practices are not markedly different from those experienced during previous financial crises.

273Anecdotally, one can think of several banks which were nationalized during the Global Financial Crisis and that, 15 years later, are still in public hands. For instance, ABN AMRO was nationalized by the Dutch government on 3^rd October, 2008 after Fortis Groups opted out of a merger deal because of its precarious financial situation.^[15] To date, ABN AMRO is the third largest banking group in the Netherlands, and the State remains one of the main shareholders with around 40 % of the shares, 15 years after the first nationalization.^[16] Another prominent example of ‘sticky’ public ownership is the Royal Bank of Scotland, which was also part of the failed ABN AMRO merger and was nationalized on 13^th October 2008, in a £37 billion bailout decision by the HM Treasury which included also Lloyd and HBOS. In 2022, the UK shareholding of NatWest Group, the successor of Royal Bank of Scotland, decreased below 50 %. This further decreased below 40 % in 2023.^[17] In March 2024 the British government ceased to be the controlling shareholder and to date holds 9.99 % of NatWest Group’s shares, committed to a full divestment by 2026.^[18]

Interestingly, these examples highlight that this is not merely a problem for States that are considered as the ‘usual suspects’, meaning countries that are commonly perceived as prone to bailouts and lax in their public finances – in Europe the so-called PIIGS.^[19] In contrast, these banks are incorporated in States that are not perceived as particularly lenient or particularly prone to fiscal interventions in the financial system. Yet, these banks are still largely in public hands, highlighting that these decisions tend to have effects for a very prolonged period of time and that finding a buyer in the market for these institutions is complex also for wealthy and stable States.

In a wider perspective, a recent IMF study tracked government interventions in financial institutions in 37 countries between 2007 and 2017. The study found systematic evidence that divestments are slow and only few countries 274managed to fully divest. The study also confirmed that it is easier to divest banks that are better capitalized, more liquid and more profitable.^[20]

The prolonged, potentially perpetual, public ownership of rescued banks limits the objectives of any effective resolution regime for distressed banks, directly and indirectly. Directly, it limits competition, increases the probability of credit misallocation and puts taxpayer money at risk; indirectly, it makes resolution tools ineffective and is prone to the strategic abuse of such situation by regulated entities.

To address this situation, regulators could completely forbid State interventions in rescuing banks. This option would represent the ‘first-best’ and is the one that regulators around the world had in mind in the aftermath of the Global Financial Crisis. However, it does not seem to be practicable. Another more modest and pragmatic approach is to re-design the regulatory framework. A framework that is able to achieve this second-best outcome should not only look at what happens before and during the State intervention (meaning at moral hazard incentives and the protection of financial stability and depositors), but also at what happens after the intervention, facilitating the effective and timely transition of the rescued bank back to the market.

This might sound like slightly more than stating the obvious, but it is not. As we will see, the current design solely focuses on reducing market distortions, moral hazard incentives and public expenditures (including the necessity for the provided aid to be ‘temporary’), but no attention is paid to those factors that – we argue – could ease the State’s divestment, thus ensuring that the precautionary recapitalization does not become a de facto bailout. The result is that, to date, the temporariness of the public ownership of a financial institution is merely virtual.

According to our model, even if in a certain measure the State divestment is subject to external factors (e. g., market instability or illiquidity, macroeconomic conditions and other factors like wars and global crises that cannot be foreseen and addressed in advance), there are some elements that an efficient resolution regime should take into account, in order to ease the State’s exit from a financial institution. Namely, these are: (i) the timing of the intervention; (ii) the solvency of the institution benefiting from the aid; (iii) the instruments employed in the recapitalization; (iv) the political interferences related to the recapitalization and the divestment.

We choose to analyze the precautionary recapitalization tool as, under the EU framework, it is the only tool that States can use to provide extraordinary pub275lic financial support to a financial institution without triggering its resolution or liquidation. In fact, in response to the massive injection of public resources in the financial sector that happened between 2007 and 2012, the EU has made public intervention in banks particularly difficult.^[21] Under the current framework, solvent banks cannot receive any public aid, with the only exception of the ‘precautionary recapitalization’ tool, which should be ‘temporary in nature’. In line with our theoretical framework, when this takes the form of capital injection, the respect of such ‘temporariness’ clause is notoriously difficult (if not impossible). This incapacity to ensure a divestment has obviously serious consequences on the credibility of the resolution system as a whole, since it undermines the basic principle that a bank in need of public support should be resolved.

3. Precautionary Recapitalization under EU Law

Under the Banking Union framework, any kind of public support provided to a distressed bank triggers the ‘failing or likely to fail’ declaration by the competent authority, thus leading to the resolution or liquidation of such bank, depending on the determination of the resolution authority.^[22] Precautionary recapitalization represents the exception to the rule. Banks that are deemed solvent (i.e., not failing or likely to fail), can apply for extraordinary public support when this is aimed to “remedy a serious disturbance in the economy of a Member State and preserve financial stability”.^[23] Such precautionary recapitalization can be in the form of State guarantees backing liquidity facilities of the central bank; a State guarantee on newly issued liabilities or the outright injection of own fund or purchase of capital instrument.^[24]

All three cases prompt the (contingent) deployment of public finance, exposing taxpayers at risk, but, in most cases, the capital injection has the further effect of shifting the control of the bank from private to public entities. For the 276purpose of this article, we mainly refer to the case of the injection of own funds or purchase of capital instruments for the analysis.

Precautionary recapitalization represents an element of flexibility within the rigid and – as the first years of application reveal – quite ineffective resolution regime. It serves a number of purposes. First, it was considered an instrument which could ease the transition from the bailout to the bail-in system, accommodating for the differences in the banking sector across the Member States; second, even under a credible bank resolution regime, public financial support could have been needed during periods of financial instability;^[25] third, precautionary recapitalizations could work also as an early intervention measure, reducing the costs arising if the financial situation further deteriorates, potentially representing a preferable option compared to late liquidation of resolution that could be far more expensive, economically and socially.^[26]

Accordingly, the European co-legislators left the wording of the provisions vague, so that implementation could be adapted to the peculiarities of the national markets and the (macro) economic conditions of the single case. This vagueness, though, created a level of uncertainty in the application of the institute, which became evident as States started to misuse precautionary recapitalizations, circumventing one of the key goals of the Banking Union, which is to sever the perverse link between the bank and its sovereign.^[27] The Commission and the European Parliament aim at tackling this distortion with the proposed crisis management and deposit insurance (CMDI) framework. Section 6 assesses the merits and pitfalls of the proposals.

Before delving into the analysis of the cases and assessing the promises and pitfalls of precautionary recapitalization, it is necessary to clearly define its legal configuration. Indeed, precautionary recapitalization can be granted only if several conditions are met. These conditions represent a specific way to set the trade-off between the distortionary effects of public aid to banks and the need to have an element of flexibility within the resolution framework, and are the outcome of a political compromise.^[28] We distinguish between conditions that are generally applicable to all forms of precautionary recapitalization and conditions that apply specifically to capital injections. Among the former, we 277further distinguish between conditions that are closely related to limiting distortions to competitions and conditions that are closely related to bank recovery issues. All conditions, in spite of their nature and scope, are discussed with specific reference to the case of direct capital injection.

The first general and competition-related condition is that the aid must be necessary “in order to remedy a serious disturbance in the economy of a Member State and preserve financial stability”. There is no specific definition of ‘serious disturbance’ as the term derives from the EU State aid framework according to which State aids are generally prohibited unless a ‘serious disturbance’ in the economy justifies them.^[29] During the financial crisis, the Commission, in its acts and decisions, recognized that the contingent economic situation could represent a serious disturbance for virtually all the cases of public intervention.^[30] Nonetheless, the absence of any quantitative or qualitative indication leaves a fair degree of discretion in determining whether or not extraordinary public financial support is needed.^[31] Such discretion emerges clearly from the different breadth with which the term has been interpreted at the European level: while the Court of Justice of the European Union (CJEU) clearly defines a ‘serious disturbance’ as one that affects the national economy of the Member State, a view also shared by the Single Resolution Board, the Commission has adopted a more discretionary approach, granting aid even in situations where the economic disturbance impacts only specific regions.^[32] In doing so, the Commission’s decisions have been criticized for their lack of transparency.^[33] 278Moreover, the aid “shall be proportionate to remedy a serious disturbance in the economy of the Member State”, so that the less intrusive and fiscally cheaper form should be preferred as long as it is equally capable to remedy such serious disturbance.

The aid “shall be of a precautionary and temporary nature”. These two concepts are quite blurred. The European Commission stated that ‘precautionary’ refers to the preparation “for possible capital needs of a bank that would materialize if economic conditions were to worsen significantly”, while ‘temporary’ means that “the State should be able to recover the aid in the short to medium term”.^[34] To date, the temporariness is the trickiest condition for capital injections, since of three cases in which banks have been recapitalized with a public intervention, the public ownership of the bank lasted for several years or is still ongoing. This inability to divest, notwithstanding the willingness of the European co-legislators, is at the core of our analysis, as it affects the functioning of the resolution system as a whole. In requiring the ‘temporariness’ of the measure, in fact, (i) the provision does not provide any tool to achieve or ensure such temporariness, (ii)sicet simpliciter, it seems to assume that the market is always liquid and ready to respond to the divestment; (iii) it does not consider the relevant factors which could, in fact, ease the public divestment and thus a limiting in time the public ownership. We contend that the problem of the temporariness in precautionary recapitalizations is directly related to the way in which the recapitalization is carried out and thus, to the design of the provision and of the regulatory framework. In this sense, a precautionary recapitalization that takes place in the early stage of the crisis is more efficient and allows for a swifter divestment.

The aid “shall be conditional on final approval under the State aid framework”. The Commission approves a precautionary recapitalization only after the submission of a restructuring plan and the implementation of the necessary burden-sharing.^[35] In practice, the burden-sharing takes place immediately after the adoption of the State aid decision and always prior to the aid’s injection into the bank. This condition wants to ensure that the disruption to the competitiveness of the EU banking market is minimized. The approval follows compliance with all other substantive competition-related conditions, especially the existence of a serious disturbance in the economy of a Member State and its proportionality, the lack of unfair advantages for the ailing banks vis-à-vis 279the competitors, the temporary nature of the State intervention as evidenced in the restructuring plan attached provided by the bank.

Moving to the general conditions that are more closely linked to bank recovery, the aid “shall be confined to solvent institutions”. This represents the quintessential condition to grant precautionary recapitalization, differentiating it from any other tool of the resolution framework. The determination of the solvency is, however, not straightforward and not immune from discretionary decisions, as it involves the assessment of the value of banks’ assets which are opaque by construction.^[36] However, in order to be effective, the recapitalization must intervene really early in the crisis.^[37] An intervention which takes place when the bank crisis is so deep that it is difficult to tell apart solvency and insolvency consists of a de facto bailout which postposes the bank’s problems and prolongs its ‘zombie’ status.^[38]

Relatedly, the precautionary recapitalization “shall not be used to offset losses that the institution has incurred or is likely to incur in the near future”. This is open to various interpretations. If the notion of ‘losses incurred’ can encompass all the losses incurred prior to the stress test resulting from the last reporting period (i.e., the last published audited or verified financial statement),^[39] it is more challenging to define the future ones. This clearly relates to the necessity 280of private sector involvement not only to limit competition distortions, but also to limit moral hazard of the ailing bank. However, the issue is more complex than it may seem at first sight as the recognition of bank losses largely depends on accounting standards, bank disclosure and supervisory practices and – anecdotally – such recognition is significantly lagged in time.^[40] The wording of the law seems to include also losses that have been incurred but are not recognized. However, the practical enforceability of this clause is doubtful, as the matter is difficult to verify, especially considering the tight timeline for granting the aid. Véron defines them as the ones that the institution is likely to face before the end of the current reporting period; while this represents a valuable guideline for the regulator, it does not solve the fundamental issues discussed above.^[41]

With regard to the specific conditions for capital injection, the law specifies that the aid can take the form of “an injection of own funds or purchase of capital instruments”. This phrasing refers to Common Equity Tier 1, Additional Tier 1 and Tier 2 instruments, as defined and regulated in the Capital Requirement Regulation.^[42] As better detailed in Section 4, practice shows that in two cases out of three in which precautionary recapitalization has been implemented, the State used a mixed percentage of shares and hybrid instruments considered as CET1, while in one case it purchased only ordinary shares.^[43] Nonetheless, the main function of precautionary recapitalizations is to address the capital shortfall by injecting high-quality capital in the ailing institutions.

The aid must be granted “at prices and on terms that do not confer an advantage upon the institution”. In interpreting this rule, the Commission referred to ‘undue advantage’ as “an advantage incompatible with the internal market under State aid rules”.^[44] Nevertheless, it should be noted that any ‘undue advantage’ can be controlled using the level of remuneration for the aid (which should align with the requirements under State aid rules) and the depth of the bank’s restructuring. Therefore, the ‘advantage’ is inherent to the construction as the remuneration for the State is necessary lower than the premium that private 281market participants would have charged. However, one interpretative solution is to consider the advantage ‘undue’ only if it is not compensated in terms of private sector involvement in losses. In this regard, Article 59(3)(e), expressly lists precautionary recapitalizations among the cases in which the resolution authority can convert or write down capital instruments outside of resolution.

Finally, the capital injection form of precautionary recapitalization can be authorized only if a stress-test or asset quality review highlights a capital shortfall. The law does not specify if such a shortfall should materialize in the baseline or stressed scenario. This additional requirement is crucial and tightly relates to the solvency requirement. In a world in which banks are perfectly transparent, the value of their asset is perfect, and the stress-tests are exactly calibrated, this requirement would ensure that only solvent but undercapitalized banks receive capital aid early on as a going-concern precautionary tool. However, this scenario rests on a number of assumptions which do not match the reality we live in, so that the signal provided by capital shortfall highlighted in the stress-test can be quite noisy and fails to capture effectively the banks that are in the condition such that precautionary recapitalization is welfare-improving.

This latter observation tightly relates with a broader reflection on the temporariness of the capital injection form of precautionary recapitalization. Unlike the other conditions discussed in this section – which are, albeit imperfectly, verifiable at the time the aid is granted – this requirement is ‘prognostic’ in nature. Furthermore, one might question the conceptual coherence of deeming aid ‘temporary’ when it takes the form of perpetual capital instruments that can be unwound only with the contractual agreement of a third party. There is an additional level of uncertainty impacting on the enforceability of the ‘temporary’ clause which is the macroeconomic conditions at the time in which the public participation should be dismissed. These problems promptly materialized in the three cases in which precautionary recapitalization was granted, as detailed in the coming Section.

4. Precautionary recapitalizations in practice

To date, precautionary recapitalization has been implemented three times: in 2015 to recapitalize two Greek banks – the Piraeus Bank^[45] and the National 282Bank of Greece^[46] – and in 2017 to recapitalize the Italian bank Monte dei Paschi di Siena.^[47] Despite heterogeneous schemes in implementing the recapitalizations, the divestment phase followed a similar timeline for all banks: after several years of substantial immobilism, divestments paced up following the raise in the interest rates in 2023. In particular: (i) Piraeus Bank has been fully divested in March 2024;^[48](ii) similarly, National Bank of Greece has been almost completely privatized in October 2024;^[49] and (iii) in November 2024, the Italian State managed to sell almost all its stake in Monte dei Paschi di Siena, but retains still an 11,7 % stake in the bank.^[50] These divestments took place almost ten years after the original recapitalization. Although some claim that these times are physiological, in reviewing the cases, this Section focuses on the enforceability of the ‘temporary’ clause. The results of the analysis will then be discussed in Section 5.

5. Critical review of the relevant factors

We saw that precautionary recapitalizations are challenging under several aspects: on the one hand, they represent an important element of flexibility in the strict resolution framework set by the Banking Union; on the other hand, they are easily used in an opportunistic way by National States, can easily distort competition and create moral hazard problems if not correctly implemented.

This ambivalent nature also reflects on the current legislation, which is mostly ‘backward oriented’ and focuses mainly on the limitations under which the aid can be provided (apparently, even the CMDI proposal), in an attempt to limit the use of public funds and thus moral hazard. Yet, if the aim is to avoid alteration in competition too, the juridical problem is wider and should also be consider why the divestment is so complicated.

On the point, data shows that, in many countries, public asset holdings in individual intervened banks remain significant even a decade after the original 294intervention.^[100] This is not a secondary aspect since: (i) public interventions come with a huge amount of taxpayers’ money; (ii) they interfere with market functions, distorting the signaling function of assets prices and financial flows; and (iii) State-owned banks are often sensible to political interference and pursue other objectives than value maximization, leading to lower profits and to an excessive risk assumption.^[101]

This Section analyses the relevant factors that could ensure a better application of this tool, easing the disposal of the State participation in the bank, whether with a merger or not and thus ensuring the temporariness of this instrument.

6. Can the CMDI proposal help?^[129]

Among the many reform proposals, the CMDI attempts to reform and strictly codify the cases for ‘extraordinary public financial support’ by adding a new article 32c to the BRRD. This reform concerns also precautionary recapitalizations, whose discipline is significantly changed with a currently very chaotic formulation.

The Commission proposes to change the discipline of precautionary recapitalization with regard to (1) the conditions to grant precautionary recapitalization; (2) the available capital instruments; (3) the enforceability of the temporary nature of the aid and the consequences of a breach of such temporary nature.

There is no substantive innovation in the conditions to grant precautionary recapitalization, as it would still be available only for solvent institutions. However, the proposal tries to specify the concept of ‘solvency’ in the context of precautionary recapitalization.^[130] To consider an institution ‘solvent’, it should not have been declared failing or likely to fail and it should not become failing or likely to fail if the aid is not granted.^[131] Moreover, the competent 303authority should conclude that no breach has occurred, or is likely to occur in the twelve following months, of any of the capital requirements provided under the European law, including the institution-specific Pillar 2 Requirements.^[132]

This reform aims to anticipate the moment when precautionary recapitalization is granted, making sure that it is truly a precautionary measure and not a de facto bailout. This attitude is to be welcome. However, the formulation and the structure of the norms are terribly convoluted and the conditions to ascertain solvency should be simplified to ensure their exact application. Moreover, a more subtle hurdle of this reform is that it is doubtful whether banks will have incentives to apply early on for precautionary recapitalization, as they may avoid acknowledging problems soon enough and delay interventions trying to force a laxer interpretation of the norms, similar to what has happened so far.^[133]

Concerning the type of instruments which should be used in carrying out the recapitalization, the Commission proposes to ban the acquisition of CET1 instruments, clearly favoring hybrid contingent convertibles.^[134] However, if the shortfall is highlighted in the adverse scenario of a stress test, the injection of CET1 instruments is still allowed.^[135] In any case, the amount of acquired CET1 instruments cannot exceed the 2 % of the institution’s risk-weighted assets. Such change aims “to ensure that the support remains temporary in nature”.^[136] Based on the analysis of this article, it is unclear whether these rigid rules can achieve their goal. In fact, there is no direct link between the type of instrument used in carrying out the recapitalization and its temporary nature.^[137]

304Finally, concerning the enforceability of the temporary clause, the Commission requires “stronger and more explicit requirements on determining in advance the duration and exit strategy for the precautionary measures”.^[138] In particular, a predefined exit strategy should be approved by the competent authority and such strategy should include a clearly specified divestment date.^[139] This seems to crystalize the current practice, making it mandatory to establish a divestment date as it is already the case for MPS. The real change relates to the consequences of the breach of such requirements. If the support is not terminated by the divestment date, the bank should be considered failing or likely to fail and be resolved or liquidated. If the support measures are not redeemed, repaid or otherwise terminated pursuing the terms of the exit strategy, the institution should be deemed as failing or likely to fail and thus liquidated or resolved. This dire consequence seems to directly descend from the fact that public financial support is a sufficient condition to declare the bank failing or likely to fail, unless it is temporary in nature.^[140] Failing to divest by the established date would de facto integrate this condition. However, this represents a rather static understanding of banking activities and banking regulation and is prone to severe unintended consequences.

First, setting a hard deadline risks generating self-fulfilling prophecies. Close to the divestment date, counterparties will be unwilling to acquire capital instruments in a bank that is about to be declared failing or likely to fail or they will apply a significant discount to the price of these instruments. Similarly, if hybrid capital instruments should be redeemed and substituted by others of at least the same quality, private investors may disproportionately increase the cost of capital.

Second, and relatedly, this approach does not take into account adverse material changes, such as macroeconomic instability, wars or natural disasters that have nothing to do with the resilience and solvency of the single institution. For instance, the initial deadline for the divestment of the public stake in MPS was set at the end of 2021 which happened to be amid the COVID-19 pandemic.

This approach assumes that the market is always liquid, meaning that there will always be a buyer for (the capital instruments of) solvent institutions. However, this has been disproven during the Global Financial Crisis and by the following academic literature.^[141] The result of this rigidity may well be to force 305into resolution or liquidation solvent banks. This would further imply that the public financial support would be subject to the rules governing liquidation and resolution. Since capital instruments are deeply subordinated, this is likely to bring about losses for taxpayers.

To avoid all these absurd unintended consequences, ex-post extensions or formulation of the initial conditions are likely to happen. Under the current formulation of the proposal, these adjustments seem to be allowed. However, ex-post renegotiation would make the credibility of the temporary clause fade, so that the temporary nature of the recapitalization would still be unenforceable.

To conclude, the CMDI proposal would transform precautionary recapitalization from an instrument of flexibility to a rigid instrument that has no actual use to address the crisis of solvent but undercapitalized banks. The reform makes some steps forward in the definition of banks’ solvency, despite the bad and convoluted formulation, and the early timing of the intervention. However, it does not consider the other key junctures that made the current precautionary recapitalization regime ineffective, including the proper design of the recapitalization instrument, and the political economy consideration.

On 24^th April 2024 the European Parliament adopted the resolution P9_TA(2024)0327, “on the proposal for a directive of the European Parliament and of the Council amending Directive 2014/59/EU as regards early intervention measures, conditions for resolution and financing of resolution action (COM(2023)0227 – C9-0135/2023–2023/0112(COD))”.^[142]

Following the CMDI proposal, the European Parliament most relevant amendments aim at (i) further specifying the conditions under which the aid can be granted; (ii) designing an effective way to ensure the temporariness of the aid.

Referring to point (i), the Parliament intervenes in specifying the requirements for the intervention, stating that the aid shall be directed “to remedy a serious disturbance in the economy of a Member State of an exceptional or systemic nature and to preserve financial stability”. With this amendment, the European Parliament aims at reducing the uncertainties highlighted also in the present paper in assessing the magnitude of the ‘serious disturbance’ which allows for a precautionary recapitalization. This amendment does not seem too relevant, as it states that precautionary recapitalization is not the ‘common tool’ to deal 306with financial crises and it burdens Member States with the obligation to justify the necessity for the aid.

Further, the Parliament tries to address the assessment of the institution’s solvency, stating that, in order to be eligible for extraordinary public financial support, an institution is considered solvent if, “based on current expectations” the competent authority “has concluded that no breach has occurred, or is likely to occur in the 12 following months” of the regulatory requirements. With such amendment, the Parliaments aligns this judgment with the Supervisory Review and Evaluation Process.

Under point (ii), the Parliament amends the CMDI in order to address several shortfall highlighted during these years. Following the approach already used in the CMDI, the European Parliament requires that, for the aid to be granted, the Member State should submit, jointly to the State aid notification and the restructuring plan, a “strategy to exit the support measure”. Such strategy should clearly state a “termination date, sale date or repayment schedule for any of the measures provided”. In order to prevent the problems already incurred in the MPS case, and outlined in paras 4.2 and 6.1, the amended CMDI proposal requires that those dates “shall not be disclosed until one year after concluding the strategy to exit the support measure, or the implementation of the remediation plan, or the assessment under the seventh subparagraph of this paragraph”. This amendment does not seem to be particularly innovative, but it is relevant insofar it transposes into the primary legislation a confidentiality duty.

The final, and perhaps most interesting amendment relates to the consequences of a breach of the exit strategy plan. If, under the CMDI proposal, failing to comply with the divestment plan entailed ipso facto a condition for the ‘failing or likely to fail’ declaration, the European Parliament seems to have a less strict, and perhaps more realistic approach. Under the amended version of the proposal, if after a precautionary recapitalization the institution is unable to comply with the predetermined exit strategy, the competent authority should request the submission of a “one-time remediation plan”. Such remediation plan “shall describe the steps to be taken in order to maintain or restore compliance with supervisory requirements, the long-term viability of the institution or entity and its capacity to repay the amount provided, as well as the associated timeframe”. In fact, this is what happened in the Monte dei Paschi di Siena case, when – after failing the first deadline – the Italian Republic asked for an extension of the plan, undertaking additional measure to ensure the compliance with the State aid measures and with the new deadline. If the second deadline is also missed, or if the competent authority deems that the one-time remediation plan is not credible or apt to restore the institution’s long-term viability, “an assessment of whether the institution or entity is failing or likely to fail shall be conducted in accordance with Article 32”.

307Such approach certifies the current practice and seems more reasonable and flexible than the one adopted by the Commission in the original CMDI proposal, as it leaves some margin for addressing external factors (e. g., global pandemics, wars, or market illiquidity) which are not under the financial institution’s control. In addition, having a ‘second chance’ can prevent the risks showed in para. 6.1, concerning ex-post renegotiation following the original formulation of the CMDI proposal. Under the amended version, in fact, it is the system itself which provides for the possibility of an ex-post renegotiation, acknowledging that there are circumstances under which a recovering bank can fail to meet its obligations due to external factors.

7. Conclusions

This contribution analyzed the causes and implications of the State’s challenges in reintegrating rescued banks into the market. We established a novel theoretical framework for precautionary State interventions in banks and, specifically, for precautionary recapitalization under EU Law. This framework served as the basis for scrutinizing the practical implementation of the measure and identifying relevant factors influencing precautionary recapitalizations. Furthermore, we conducted a preliminary analysis of the impact of the CMDI proposal original version and of the CMDI as amended by the European Parliament.

The existing framework, and unfortunately, the CMDI proposal, predominantly adopts a ‘backward oriented’ approach, primarily focusing on the constraints for providing aid to limit moral hazard. While this is necessary and praiseworthy, it completely overlooks the costs brought about by the events following the intervention, in particular the ‘stickiness’ of public ownership in the rescued banks. On the contrary, the amended version of the CMDI proposal seems to be more flexible and reasonable, but it is still extremely vague and insufficient in assessing whether an institution is solvent or insolvent. In addition, in several parts it reiterates the same approach (thus, the same criticalities) of the original CMDI proposal. Accordingly, a definitive solution to ensure a smooth re-entry of banks into the market is still elusive. Notably, the nature of the capital instrument (ordinary shares or hybrid instruments) appears to have minimal inherent impact on the State’s divestment prospects. Instead, the design of the capital instrument, such as the use of instruments like reverse contingent convertibles, or other instruments specifically designed and tailored on the peculiarities of the single distressed institution, emerges as a more effective strategy.

Another critical determinant of a successful market re-entry is the timing of recapitalization. Precautionary recapitalizations prove most effective when ap308plied to solvent institutions early in a crisis. However, practical implementation faces challenges, as banks are incentivized to delay loss recognition, and distinguishing between solvency and liquidity crises in distressed financial institutions is complex. Moreover, the mere request for precautionary recapitalization can worsen the bank’s position, creating a self-fulfilling prophecy. In general, banks failing stress tests or equivalent assessments under baseline scenarios should be ineligible for precautionary recapitalization. For these reasons, we propose that primary legislation should be amended in order to (i) vest the SSM with the power to act preventively in the crisis and (ii) grant the SSM a broad discretion in deciding and tailoring the instruments of the recapitalization. None of the proposals takes into account such aspect, though.

Our findings underscore the need for a ‘forward looking approach’, combining well-designed capital instruments and timely interventions tailored to the unique circumstances of each financial institution: clear legal mandates and regulatory discretion are crucial for addressing the intricacies of the market re-entry process in the aftermath of banking rescues.