The Covid-19 pandemic will leave deep scars across the globe, particularly in the euro area. Given that the health of banking systems is inextricably tied to the performance of the underlying economies, the non-performing loans (NPLs) of banks are an important issue. What are the policy options to safeguard the integrity and functionality of the banking system? And what are the criteria defining the desired response? This column will address these questions in the context of the EU.
What makes the identification of a suitable policy response particularly difficult is the strong reliance on banks and the apparent ‘overbanking’ in Europe (Pagano et al. 2014). At the national level, banking markets are highly concentrated, and many institutions are considered too-big-to-fail. The structurally low profitability of European banking makes this even more of a concern.
Policy responses should take these structural issues into consideration. In particular, the policy actions should neither reinforce the substantial reliance on banks, nor perpetuate a ‘legacy banking’ architecture that is nation-centric and prone to a ‘doom-loop’ between the fiscal state of national governments and the state of the banking system. The extent to which financial markets could play a more prominent role should also be considered.
In this column, we discuss and evaluate a variety of policy options that are considered in the current debate on how to deal with potential problems of the European banking sector, amplified by the Covid-19 crisis.1 The evaluation is based on a set of criteria that, in our view, capture the effectiveness and credibility of a proper policy response.
Evaluating policy options
We carefully distill the requirements that a desirable policy response should meet in light of two objectives of the policy intervention: (i) the stability of the banking system and (ii) the ability of the banking system to fulfill its important role in society. Subsequently, we apply the criteria to the policy options.
We define the following five criteria:
1. Effectiveness: Can the overall objectives be achieved? Does the option deal effectively with the problem at hand? Does the implemented model make a difference?
2. Feasibility: Is the option feasible in a broad sense? We consider various dimensions regarding feasibility:
a. Is the option feasible (e.g. not rejected outright by the legislative process or by the treasuries involved)?
b. Is a political mandate for the option possible?
c. Are policymakers and regulators able to execute the policy?
d. Is the option viable in a narrow sense (i.e. not too complex)?
3. Credibility of the policy: If put in place, can it be carried through over time?
a. Is the problem of ‘regulatory capture’ addressed?
b. Is the option resilient to the ‘too-many-to-fail-problem’ of policymaking?
c. Is the policy time consistent, in the sense that incentives to re-adjust are kept at bay so that ex post credibility can be ensured?
4. Alignment with incentives of private players: Does the (public) intervention leave ex ante the right incentives and initiative for banks and firms? This question includes the following aspects:
a. Does the option prevent ‘zombification’ of firms and/or banks?
b. Can regulatory arbitrage be contained?
c. Does the option allow for private initiatives to deal with problems at hand (e.g. no weakening restructuring incentives)?
d. Is the flow of credit to firms, in particular SMEs, sustained?
5. Structural impact at the bank level: This criterion assesses the impact of the policy on the longer-term challenges of the banking industry. The following questions are considered:
a. Does the option respond to the overbanking issue?
b. Does the option limit the market power of established institutions?
c. Is this fostering positive renewal in the financial system?
d. Does it strengthen the role of capital markets in Europe?
We evaluate the various options along the criteria defined above, giving an assessment based on a three point-scale: yes (green), medium (yellow), and no (red). Table 1 summarises the results of the evaluation. For more details see Boot et al. (2021).
Option zero: Private recapitalisation
The basic option to improve the resilience of a single bank is to have it strengthen its capital base. This would allow the bank to deal with the problems at hand on its own (e.g. raise equity and manage non-performing loan). Such private sector initiatives seem preferable, if circumstances permit. A privately recapitalised bank would take full control over its own destiny and have the right incentives for making appropriate business decisions. As Table 1 indicates, this option is effective: it provides the right incentives to private players and would have the right structural impact at the bank level. Feasibility might be an issue, but it is a useful benchmark.
Option 1: Forbearance
The forbearance we focus on aims at giving banks some leeway in meeting regulatory requirements. Banking supervisors have stressed their willingness to accept temporary breaches of regulatory capital requirements if the shortfall is due to pandemic-related provisioning. The actions taken include deactivating surcharges for systemically important banks, lowering risk weights, excluding assets when calculating the leverage ratio, and temporarily suspending newly introduced accounting rules (IFRS 9). We evaluate option 1 under the assumption that it is the only measure taken by authorities and interpreted broadly (including, for example, relaxing accounting rules). As highlighted in Table 1, this option is feasible, but it would be largely ineffective because it does not contribute to the overall stability of the financial system in the medium term. Forbearance may beg for more forbearance, as often happens. Financial institutions will learn the lesson that tough regulatory rules will be bent whenever the risk outlook is gloomy enough. This option might provide perverse incentives to private players as a sole measure. In particular, it could entail substantial elements of the banking union to be suspended. It may lead to the zombification of banks and encourage continued lending to sclerotic firms. Moreover, this option would not have a favourable impact on the banking sector.
Option 2: Recapitalisation via public money
Recapitalization by governments refers to precautionary and mandatory recapitalisations, as suggested by Schularick et al. (2020). For European banks, the authors estimate a capital shortfall between €60 billion and €600 billion, depending on the pandemic scenario. The acceptance of such publicly financed equity infusion would be compulsory, conditional on not passing a stress test. Hence, this alternative could build on the previously discussed basic option if recapitalising via private money cannot be accomplished. As reported in Table 1, this option is effective and credible, but it would be difficult to implement because it encounters resistance as it involves public money. Moreover, this option could perpetuate overbanking. To contain this risk, it is key to attach strong conditions to the public infusion of capital.
Option 3: De-risking via asset sales
In case a bank is unwilling or unable to raise new equity, it may seek to improve its capitalisation ratio by selling assets, or more generally, by ‘de-risking’. This can lead to fire sales, producing systemic risk. As shown in Table 1, this option is feasible and would have a structural impact to the banking system, but it is not effective because the option might imply fire sales and threaten wider financial stability. Moreover, banks would have a strong incentive to substantially cut back on lending. This option is not credible because it may contribute to systemic risk in the economy, which, in turn, may make subsequent bailouts more, rather than less, likely. In terms of alignment with private incentives, this option would limit zombification and contain moral hazard.
Option 4: Asset separation through an individual bad bank model at the national level
‘Bad banks’ – also called asset management companies – have been used in the past to resolve calamities in banking. Sweden in the early 1990s is an interesting case in point. A bad bank typically needs substantial support by the government because losses will be realised. Creating a bad bank tends to invite challenges that relate to the nature of the assets transferred. What is a fair price of a particular loan? How can policymakers ensure that the private information owned by the bank is shared with the management of the bad bank? How can the bad bank optimally manage the assets and maximise recovery? In our assessment we assume that there are clear incentives to maximise the value of recovery. As reported in Table 1, this option would be effective and would provide the right incentives if the bad bank is managed and incentivised in the right way. The risk of moral hazard is limited, but there are still risks concerning national champions. However, in terms of feasibility this option is complex. The structural impact on the banking sector is constrained if national banking champions are preserved or created.
Option 5: Asset separation through an EU-wide bad bank model
A variation on option 4 is to set up an EU-wide bad bank – the irrevocable transfer of non-performing loans by banks to a supranational asset management company. Collecting these loans from across the single market, as reported in Table 1, is effective and might be managed more efficiently and objectively than national vehicles. However, complexity substantial undermines feasibility. Moreover, the supranational solution may trigger ‘mutualisation’ concerns. That being said, it might also incentivize the creation of a market for distressed assets, and ultimately help the development of the Capital Markets Union (Beck 2017). In doing so, it could have a favourable structural impact on the banking system in Europe.
Option 6: Loss capping through debt restructuring/conversion
A different approach to tackling NPLs is a partial transfer of the default risk on each bank loan from a bank to (typically) a public authority. One avenue is via a targeted insurance scheme that offers coverage for realised losses exceeding a threshold level. Another avenue is a scheme where a public body refinances existing bank loans (where these new loans have limited recourse on the bank involved). It should be noted that this is effectively a capital infusion by the government since the bank receives more than the distressed value of the loan. As highlighted in Table 1, this option would be effective and, compared to the bad bank option, it leaves incentives more aligned, preserves the informational advantage of bank relationships, yet might have weaker collection incentives. The feasibility largely depends on the presence of an effective state-owned ‘development bank’, or a similarly established institution. Credibility relies not just on the effectiveness of the development bank, but also on the political strength to deal with the too-big-to-fail problem and moral hazard (see option four). The restructuring consequences of the scheme are partial, as individual loans are the focus; the risk of preserving national champions still exists.
Table 1 Overview: Criteria-based assessment of selected options
In this column, we have highlighted and assessed several policy options that aim at improving the resilience of European banks. The policy options range from forbearance, public recapitalisation, asset sales/de-risking to asset separation (bad bank at national or EU level), and loan conversion by state banks. We have evaluated each along a list of five criteria that should define the desired response: effectiveness, feasibility, credibility of policy, alignment with private incentives (mitigating moral hazard), and structural impact on the banking industry. Clearly, none of them are a panacea and there are positive and negative aspects to all of them.
Our assessment indicates that asset separation and loan conversion might be crucial for the viability of the European banking system, as these options dominate the other three (forbearance, public re-capitalisation, and de-risking). Among the preferred options, we point to the benefits that an EU-wide bad bank might bring (option 5). Such EU-wide system, collecting NPLs from across the Single Market, could potentially operate more efficiently and could be less prone to capture than national vehicles. On the downside, informational advantages embedded in the long-term bank-firm relationships might be lost. This loss of information would be avoided in a debt conversion scheme (option 6), and possibly also with a national bad bank (option 4).
When it comes to feasibility, the national bad bank and/or debt conversion options might have a benefit but have a disadvantage when it comes to the credibility of the policy. National authorities might still find themselves ‘captured’ by domestic banks.
In this assessment, we have taken no stance on whether an infusion of public money is easier to accomplish at the national or at the European level. We also have not attached a value to the risk transfer to the European level with the pan-European bad bank. It could improve risk diversification but, for sure, raise mutualisation concerns. Although we do not make a specific recommendation, we provide a framework for policymakers to guide them in their decision making.
Author’ note: An extended version of this column is available as SAFE White Paper No. 79.
Beck, T (2017), “An asset management company for the Eurozone: Time to revive an old idea”, VoxEU.org, 24 April.
Boot, A, E Carletti, H-H Kotz, J P Krahnen, L Pelizzon and M Subrahmanyam (2020), “Corona and Financial Stability 4.0: Implementing a European Pandemic Equity Fund”, VoxEU.org, 22 April.
Boot, A, E Carletti, H-H Kotz, J Krahnen, L Pelizzon and M Subrahmanyam (2021), “Corona and banking - A financial crisis in slow motion? An evaluation of the policy options”, SAFE White Paper 79.
G30 (2020), “Reviving and Restructuring the Corporate Sector Post-Covid: Designing Public Policy Interventions”, report of the Working Group on Corporate Sector Revitalization, G30, Washington DC.
Pagano, M, S Langfield, V Acharya, A Boot, M Brunnermeier, C Buch, M Hellwig, A Sapir, and I van den Burg (2014), “Is Europe Overbanked?”, Reports of the Advisory Scientific Committee 4.
Schularick, M, S Steffen and T H Tröger (2020), “Bank capital and the European recovery from the COVID-19 crisis”, SAFE White Paper 69.
1 As emphasised in a recent G30 report (G30 2020) and our related work (Boot et al. 2020), a focus on the resilience of the business sector is also important and helps limit loan losses imposed on banks.