Thorsten Beck

Authors

Thorsten Beck^[1]Professor of Banking and Finance, The Business School (formerly Cass), City, University of London, Director, Florence School of Banking and Finance, European University Institute, and Research … Continue reading

 

Unlike during the Global Financial Crisis and Great Recession, the financial sector has not been at the core of the current crisis. Rather, the financial sector has been affected as much as other sectors by the public health crisis and the lockdown measures imposed by governments. Borrowers affected by the pandemic are less likely to repay loans and the lowering of interest rates across the globe has put pressure on banks’ interest margins.

At the same time, however, the financial sector has served a critical function in the transmission of multiple support measures of governments and central banks to limit and mitigate the economic fall-out from the pandemic. Specifically, monetary authorities have not only reduced interest rates (where they were not already in negative territory as in the euro area), but also expanded asset purchase programmes and stepped in as market maker of last resort where financial markets showed clear disruptions. These aggressive monetary policy actions have had the objective to maintain liquidity and credit to the real economy.

There has also been a wide range of government support programmes, including (i) compensating firms for the containment measures enforced to close businesses or reduce economic activity such as government-sponsored job retention programmes paying firms for specific fixed costs such as rents or interest on loans, (ii) tax cuts or holidays, and (iii) public guarantee schemes and moratoria on loan repayments. Payments of support programmes are transacted through the banking systems and guaranteed loans often granted by banks.^[2]See ESRB (2021) for more detail on support measures in the EU.

Finally, there has been a variety of supervisory measures, including (i) capital relief (i.e., allowing banks to operate below regulatory minimum thresholds), (ii) relaxation of loan classification and provisioning rules and, (iii) (in the euro area) delay of stress tests and the supervisory review and evaluation process (SREP) to 2021. These actions aimed at easing operational pressure on banks and providing incentives to maintain if not extend lending during the crisis. In return, banks were requested to constrain profit distribution to thus maintain the necessary liquidity and buffers both for lending and for loss absorption.

These different support measures can also – at least partly- explain why banks have not suffered as one might have expected given the economic downturn. On the one hand, banks have benefitted from higher fee-based revenue from activities in financial markets.  On the other hand, loan loss recognition has been delayed, an effect that seems stronger than the effect of loan moratoria. And while lending might not have increased by as much as simple multiplier models of the capital relief suggested, buffers have been maintained if not built up and resilience strengthened, even if at the costs of lower returns for shareholders (Hardy, 2021).  This also implies, however, that the pain might still be ahead.

 

The real economy after the pandemic

As much as governments have provided emergency support for real economies across the globe and thus taken on economic losses stemming from the pandemic, there is likely to be a fallout.  Specifically, while government support has achieved to ‘freeze the economy’ and avoid unnecessary frictions of illiquidity and insolvency in the real economy, this has also put on hold the market-based process of resource allocation. And as a lot of support has come in the form of debt rather than grants, many firms might exit the crisis with a debt overhang.

A first important step will be the exit strategy from support for the real economy, both in timing and in speed.^[3]For a more extensive discussion on exit strategies, see Beck, Bruno and Carletti (2021) At the core lies the tension between “Keynes” and “Schumpeter”. On the one hand, continuous support even beyond the opening-up phase can be justified with the attempt to avoid hysteresis, i.e., the risk that the current severe economic downturn and consequent high unemployment (in absence of support measures) cause unemployed individuals to lose their job skills or become demotivated, turning into high rates of long-term or structural unemployment. Such scaring effects would hamper not only economic recovery but also permanently reduce potential output and ultimately result in lower long-term growth rates.   Similar arguments can be developed for other economic input factors, such as commercial real estate and manufacturing capital. This is not only challenging from macroeconomic perspective, but also from social and political viewpoints. Supporting firms and people is thus the first priority – and through such support, pressure is also being relieved on banks.

On the other hand, the pandemic will have (possibly permanently) changed the returns on activity in different sectors and industries. There is thus a need for reallocation of resources within the economy post-pandemic.  This requires a process of Schumpeterian “creative destruction”, where some firms, even if viable before the outbreak of the pandemic, may have to undertake a profound transformation towards new products, services and/or markets, and new firms are created in sectors and industries with growth opportunities. Such a process would be impossible, if support measures keep all firms in their current structure alive, independent of whether they are viable in their current structure in the long-run or not. Capital and labour would be tied in such firms, reallocation thus impossible and growth depressed.

At the core of this tension is uncertainty.  While Europe has been emerging from the third wave, it is not clear whether this will be final one. While the introduction of different vaccines has provided hope, their effectiveness against further mutations is unclear as is the point when COVID-19 is no longer to be regarded as pandemic but limited to local and possibly much less fatal outbreaks. Given the uncertain trajectory of the (exit from the) public health crisis, there is similar uncertainty about the necessary constraints on socio-economic life, which will impact the economic recovery. On the one hand, this speaks for maintaining the support for longer until the recovery process has clearly taken off, thus also avoiding cliff effects that can result in wide-spread insolvency and unemployment; on the other hand, this calls for a more differentiated approach in support going forward, focusing on sectors that are most affected by continuing constraints on economic activity and where persistence effects in consumption will imply a slower recovery process. Most importantly, however, this calls for erring on the side of maintaining support for too long rather than terminating too early.

On a more macroeconomic level, these considerations also strongly speak against repeating the mistake from the early 2010s when an expansionary fiscal policy to (successfully) mitigate the extent of the Great Recession (or rather: prevent a second Great Depression) was quickly replaced by an austerity stance on both sides of the Atlantic – in the US due to political conflict between president and Congress and in Europe to comply with arbitrary fiscal policy constraints and the political desire of several euro area core countries to lead periphery countries with ‘good example’ on how to implement austerity. On the euro area level, this ultimately resulted in a deflationary fiscal policy stance, deepening the economic recession and putting too much burden on monetary policy. While one might argue about the appropriate size of fiscal policy stimulus (a discussion primarily on-going in the US), it would be economically illiterate and damaging to use the inadequate appeal to ‘household finances’ to ‘recover’ government expenditures incurred during the crisis and aggressively reduce government deficits and debt levels. As the example of austerity in the UK in the first half of the 2010s has shown, this can throttle a speedy recovery, augment deficits and debt levels further, and have severe socio-political repercussions.

It is clear, however, that as we proceed towards an exit from the public health crisis and thus, towards broader-based economic recovery, the weights on the reallocation process in the real economy become stronger compared to the weights on the survival/hysteresis arguments. Some sectors that rely a lot on personal interaction or physical presence will have to shrink, while others that rely on remote interaction will have growth potential. There might also be a geographic reallocation of growth potential, possibly away from larger cities. On the micro-level this implies that some firms are no longer viable while there is the potential for new enterprises entering growth sectors.

This reallocation process will not necessarily be without frictions. Important will be the distinction between (i) unviable firms and (ii) viable but overindebted firms, where among the latter some might already have entered the pandemic with overleveraged balance sheets, while others have seen an unsustainable increase in debt during the pandemic. The regular insolvency framework might not be appropriate for widespread corporate fragility nor might be its heavy focus on liquidation rather than debt restructuring.

There are different ways to address widespread corporate fragility (Sandbu, 2020): one would be to convert emergency loans – either direct ones or bank loans guaranteed by the government – into grants; however, this would be costly and would probably be mis-targeted, as it would benefit firms that might not rely on such support while keeping alive unviable firms. A more targeted measure would be government equity support for viable but overindebted firms; however, this will be difficult to manage given the large number of firms and the limited if not negative track record of governments to pick winners. A third option would be a bank-based restructuring process, as especially for smaller firms in Europe the largest part of their debt will be bank loans, so that banks have the right information and capacity to restructure debt. The main problem is whether banks have the right incentives to undertake this role in the societally most efficient way; if they provide too much debt relief to benefit from future relationships with their clients, borrowers might jump ship to other banks afterwards; if they provide too little, the economy might end up with walking zombies, even though these clients are tied to the bank, deteriorating banks’ asset quality. Regulatory rules (as well as taxation) might influence banks’ actions. Having a central role for banks in this process, however, might also divert their resources from the necessary funding of new companies and thus the economic recovery process.

In previous crises, this challenge has been addressed with asset management companies (AMC), which can help reduce non-performing assets on banks’ balance sheets by transferring them to special purpose vehicles. Among the benefits of AMCs are economies of scale in the workout of non-performing assets and helping to close the gap in pricing, when asset prices are temporarily depressed. AMCs might also be in a better position to restructure the debt of borrower with multiple bank relationships and – by taking on a coordination role – avoid fire sales that result in a further depression of asset prices. At the same time, being able to off-load non-performing assets allows banks to focus on lending to performing and new borrowers.  While in theory, similar effects can be achieved through market-based securitisation schemes, asymmetric information between banks and investors (resulting in a lemons problem) and the more urgent need for banks to offload assets than for investors to buy might result in market failures, in addition to absorption limits of private markets. Public-private partnerships, where publicly-supported AMCs are partly funded by private investors, seem a more promising route.   The more successful AMCs, including after the Global Financial Crisis in Ireland and Spain, however, have dealt with real estate rather than with SME loans, which are more heterogeneous, complex and costly to work-out.

There are constraints, however, on the use of publicly-supported AMCs, as they are subject to state aid conditions and have to be compatible with BRRD and can thus only be established for solvent banks with viable business models. Further, the effectiveness of AMCs might be hindered by slow and ineffective corporate insolvency frameworks, a problem that is stronger in some EU member states than in others.

A critical issue are the prices at which AMCs take on non-performing assets from banks. If purchased at book prices, this involves a transfer of losses from banks to the AMC and ultimately government, in conflict with state aid rules. A transfer at market prices, on the other hand, can result in large losses for banks and thus the need for recapitalisation or resolution. A transfer at the economic value (most likely in between market and book values) might reduce bank losses, but at the same time result in the need for government resources to be tied up in the AMC.

 

Banks’ asset quality after the pandemic

Debt restructuring of some firms and liquidation of others will have obvious repercussions for the quality of banks’ asset portfolios. There is certainly variation across banks and countries in this negative impact. It is important, however, that these losses be recognised; any delay can result in zombie lending and further accumulation of losses as the case of Japan in the 1990s has shown. At the same time, leaving the process completely to banks creates the risk of overwhelming them and thus hindering the reallocation and recovery process. AMCs as discussed in the previous section, might come in useful here.

In spring 2020, loan loss classification standards were relaxed in Europe, with supervisors advising banks “to make use of the flexibility provided by standards and take a long-term view in assessing which creditors are in a good position to recover from the crisis.” (ESRB, 2021), while at the same time forcing banks to start accumulating general provisions in response to the deterioration of the macroeconomic scenario. Such flexibility, however, can result in opaqueness of banks’ balance sheets and provide perverse incentives for banks to roll-over loans to non-performing borrowers and thus zombie lending.  Evergreening and zombie lending has negative repercussions not only for average firm growth but also negative growth implications for non-zombie firms who might be undercut in pricing by zombie firms and who cannot expand at the expense of zombie firms. It also prevents the entry of new innovative firms that might contribute to overall (productivity) growth in an industry or sector (Adalet McGowan et al., 2018). It is thus clear that a return to forward-looking loan loss provisions is an important part of the exit strategy for regulators.

As banks have to provision for prospective loan losses, incur such loan losses, or have to adjust book value in the context of transfers to AMCs, the question on how to deal with the consequent bank fragility arises. Stress tests under way in Europe will give a clearer picture of banks’ prospective post-pandemic asset quality; the continuous uncertainty on the course of pandemic and economic recovery, however, makes clear prediction on future asset quality and the likelihood of different scenarios more difficult. Importantly, authorities have to be prepared for possible bank failures, a topic I will turn to next.

 

Bank resolution in Europe – ready for the first big test?

The absence of effective bank resolution frameworks forced European authorities in 2008 into one of two ‘corner solutions’: send failing banks into corporate liquidation processes or bail them out. The former ignores the interconnected character of banking and the negative externalities that the failure of banks cause for borrowers, depositors and the broader economy; the global shock of the Lehman Brothers’ failure illustrated these effects and can explain why European authorities went mostly for the bail-out option, at least in the case of larger banks. However, the bail-out implied not only losses for taxpayers and consequent cuts in other government budget lines (one of causes for the subsequent rise of populist parties), but also raises moral hazard concerns.

The introduction or reforms of bank resolution regimes across Europe aimed at ending such bail-outs, while at the same time allowing for efficient resolution or liquidation of failing banks and minimising negative externalities and spill-over effects on other banks and the real economy. The Bank Recovery and Resolution Directive (BRRD), translated into national legislation, created common standards, including restrictions on the use of taxpayer resources.

Even in the years leading up to the adoption of the BRRD across the EU, there was already a shift from bail-outs to bail-ins (World Bank, 2016). Most prominently, the failure of Banco Espirito Santo (BES) in Portugal was addressed by a mix of bail-in of junior debtholders, a good-bank bad-bank split and a bridge bank structure. Specifically, the resolution involved the immediate creation of a bridge bank named Novo Banco that received sound assets and liabilities such as cash, retail deposits, performing loans, and central bank funding. In contrast, shareholders and junior bondholders were bailed in and thus left with the toxic assets that led to the mounting losses, which remained in a “bad bank” that was subsequently liquidated. Importantly, the newly created bank became fully owned by the Portuguese Resolution Fund, which provided the entirety of the Euros 4.9 billion of capital. The financial resources of the Fund did not include public money, as it was financed by the initial and periodic contributions of all of the country’s lenders as well as the proceeds from a levy on the banking sector. Beck, Da-Rocha-Lopes and Silva (2021) show that firms linked to BES suffered a significant contraction of credit at the intensive margin, but were on average able to compensate for the supply-driven shock. However, affected SMEs experienced a binding reduction of funds available through credit lines, and those with lower internal liquidity increased precautionary cash holdings and reduced investment and employment. This suggests that bank resolution without bail-outs and taxpayer support can limit though not eliminate real sector costs from bank failures. It is important to stress, however, that the failure of BES was an idiosyncratic case, not related to deeper imbalances or fragilities in the Portuguese banking system. The swift intervention thus limited any contagion effects, though the idiosyncratic nature of the BES failure might have limited them anyway. One cannot conclude from these findings for this specific case, that the fallout of bail-ins would be similar in a systemic crisis situation.

Since the adoption of the BRRD, gaps in the new resolution frameworks have become clear. These gaps include a focus on liquidation for banks, for which there is no positive public interest assessment for resolution (assessment that normal insolvency proceedings would “give rise to significant adverse effects on the financial system and severely impede the functioning of the real economy in one or several Member States”, SRB, 2019). While there might be indeed no such narrowly defined public interest in the case of many smaller banks, widespread insolvency and liquidation of several smaller banks, especially if geographically concentrated can have severe negative economic repercussions (e.g., Ashcraft, 2005). Further, “significant differences in national legal regimes for the liquidation of banks imply divergences from the European supervisory framework; they generate level playing field concerns that might impair banking market integration and they may stand in the way of a smooth exit from the market for the weakest players” (Enria, 2020). Discussions on possible reforms are currently under way, but any such reforms will be too late to address possible bank fragility post-COVID-19.

Can the current bank resolution framework be used in a systemic banking crisis?  Theory is ambiguous on the effect of a more comprehensive bank resolution framework on stability during instances of systemic distress. On the one hand, reducing the likelihood of bailouts and thus taxpayer support, allowing early intervention, and providing ample tools for resolution of failing banks reduces moral hazard risk (Repullo, 2005; Farhi and Tirole, 2012). Specifically, bail-in and clarity on how losses will be distributed in case of bank failure can increase market discipline by equity and debtholders of banks. They can also reduce incentives for too high leverage on banks’ balance sheets (Adrian and Shin, 2014). On the other hand, a rule-based system that ties regulators’ hands can result in bank runs and contagion if regulators have private information about bank performance (Walther and White, 2020). Rule-based bail-ins might make banks more vulnerable to adverse events and thus destabilize the financial system in the middle of a crisis, through direct interlinkages of banks holding each other’s’ claims, as well as information effects and a sudden reassessment of bank risk (Acharya and Yorulmazer, 2008; Eisert and Eufinger, 2018). According to this view, bailouts of failing banks (which were supposed to end with the post-2008 reforms) can protect other banks from contagion and thus provide incentives to reduce risk-taking (Cordella and Yeyati, 2003; Dell’Ariccia and Ratnovski, 2019). There might also be economic costs of too rigid an application of rules, resulting in underinvestment (Keister, 2015; Leonello, 2018).

Beck, Radev and Schnabel (2020)’s empirical assessment of the relationship between bank resolution frameworks and systemic risk sheds doubt on the usefulness of bank resolution frameworks during systemic banking crises. Specifically, they show that banks in countries with more comprehensive bank resolution frameworks experience a higher increase in systemic risk contributions after system-wide shocks, such as the Lehman Brothers’ failure or the Greek debt crisis; further, these amplification effects are mainly driven by the overall bail-in framework and the tools and powers the resolution authority has at its disposal, while the existence of a designated resolution authority is related to system-wide shocks and banks’ systemic in a dampening way. Interestingly, the authors do not find such amplifying effects during idiosyncratic shocks (such as, for example, the failure of Banco Espirito Santo, discussed above).  These results suggest that more comprehensive bank resolution may exacerbate the effects of system-wide shocks and should not be solely relied on in cases of systemic distress.

The theoretical and empirical evidence matches experience from previous crises across the globe, where often blanket guarantees, system-wide recapitalisation efforts and – as discussed above – asset management companies are being used (Laven and Valencia, 2018). Bank resolution frameworks are designed for idiosyncratic failures and both the toolbox of resolution techniques and political appetite for bail-ins shrink in the face of systemic fragility, something also referred to as scale diseconomies of resolution (De Young et al, 2013, Beck, 2011). Specifically, the simultaneous failure of several institutions not only exacerbates the stress experienced by directly or indirectly affected institutions, but also limits the effectiveness of resolution techniques, such as purchase and assumption of failing banks by healthy ones, as potential acquirers might either be affected themselves or be reluctant to acquire in times of high uncertainty.

In the context of multiple and geographically concentrated bank fragility in Europe, a strict adherence to the current framework, designed for idiosyncratic bank failures (just to stress this again), might exacerbate fragility, as discussed above. A flexible approach to the use of the different tools discussed above including where a positive public interest assessment might not be met in normal times, with waivers of state aid rules where necessary and – most importantly – pan-European solutions, is critical. As a focus on purely national fiscal policy stances is no longer an option within the euro area, forcing resolution, restructuring and recapitalisation decisions onto the national level can restart the vicious cycle of bank and sovereign fragility we saw in the early 2010s. While a completion of the banking union and a reform of the BRRD is not feasible to address bank fragility in the short-run, the spirit of a complete banking union should be applied. This also implies early coordination between regulators, resolution authorities and governments on the national and European level.

 

The crisis as opportunity

While the immediate objective of the banking union was to cut the vicious cycle between bank and sovereign fragility, the medium- to long-term objective has been to create a Single Market in Banking, moving away from national towards an integrated banking system.  Neither of these two objectives has been fully accomplished.  The banking union is not complete and the early stages of the COVID-19 crisis increased fear of a renewed bank-sovereign fragility cycle, ultimately countered with the aggressive actions by the ECB and the strong signal sent by the European Recovery Fund (Next Generation EU).  One example for negative repercussions of an incomplete banking (and fiscal) union emerged in spring 2020: while the ECB asked for restrictions on profit distribution on the group-level within the EU, several national supervisors also restricted within-group profit distribution, effectively undermining the Single Market of free capital movement but with the valid argument that local subsidiaries benefit from national fiscal support packages.  And while banks in Central, Eastern and South Eastern Europe are much less dependent on parent bank funding that a decade ago, memories of lending retrenchment in the wake of the Global Financial Crisis are still fresh, while an incomplete banking union leaves national authorities in these countries in a relative weak position vis-à-vis home country authorities (Ahmad et al. 2019).

However, even a completion of the banking union is only a necessary but not sufficient step towards a Single Market in Banking. Cross-border mergers can help delink banks from countries and thus governments; but it is the same governments that often stand in the way, as the recent example of Germany has shown where the government actively tried to facilitate a merger of the two largest private banks.

Beyond creating a truly Single Market in banking, where larger banks are European rather than national, one can consider a second longer-term objective: reducing the bank-bias in the European financial system (Langfield and Pagano, 2016). Strengthening public capital markets is only one aspect, strengthening private capital markets, including equity funds, angel financing and venture capitalists are other important aspects. Balancing the financial system is critical in the context of the increasing importance of intangible relative to tangible capital (Haskel and Westlake, 2017). Recent research has shown the limitations that banking faces when enhancing growth of industries and economies increasingly relying on intangible assets that are harder to be used as collateral that can be recovered and resold and with more uncertain investment projects (Beck et al., 2020).  This is consistent with increasing evidence that such industries are more likely to be financed by non-bank financial institutions, including venture capitalists, equity funds but also through public capital markets (Dell’Ariccia et al., 2021).

Another medium- to long-term challenge for the European banking system is the rise of fintech and bigtech companies, which have the potential to disrupt banking markets. Fintech companies have undermined banks’ franchise in specific services, most prominently payment services, and are thus threatening economies of scope and scale banks have been enjoying by offering bundles of services.  Bigtech companies have a critical advantage vis-à-vis banks through their access to big data and large networks, which they can use for an envelopment strategy in new markets, including financial services. Ultimately, the competitive threat to banks from bigtechs and banks’ reactions will be critically determined by the regulatory response.

 

Conclusions

The crisis has not started in the banking system, but banks have been a critical transmission tool for the management of the economic crisis. It is clear, however, that unless the phasing out of support programmes is undertaken carefully and in a coordinated way, there is the risk that corporate distress will result in banking distress, in the form of a vicious cycle that might even bring sovereign fragility back into the picture. And while the bank resolution tools at the disposition of authorities are vastly superior to the ones available in 2008/9, it is doubtful that they are sufficient to resolve multiple bank failures, especially if geographically concentrated.

Careful coordination between different national authorities (bank supervisors, resolution authorities, and governments) and between European and national authorities is needed to not only design coordinated exit plans but also put in place the necessary plans for severe fragility in an adverse scenario; plans that build on existing frameworks, but with the necessary flexibility to address systemic banking distress.

On the upside, if properly handled, any bank fragility resulting from the pandemic and the economic fallout can be used to kickstart a deeper restructuring of Europe’s banking systems, completing the banking union and building a truly Single Market in banking in Europe. The time to prepare is now.

 

References

Acharya, V., and Yorulmazer, T. (2008). Cash-in-the-Market pricing and optimal resolution of bank failure. Review of Financial Studies, 21, 2705-2742.

Adelet McGowan, M., Andrews, D., and Millot, V. (2018). The walking dead? Zombie firms and productivity performance in OECD countries. Economic Policy, 33, 685-736.

Adrian, T., and Shin, H.S. (2014). Procyclical Leverage and Value-at-Risk. Review of Financial Studies, 27, 373-403.

Ahmad, I., Beck, T., d’Hulster, K., Lintner, P., and Unsal, F. (2019). Banking Supervision and Resolution in the EU. Effects on Small Host Countries in Central, Eastern, South Eastern Europe. World Bank Group.

Ashcraft, A. B. (2005). Are banks really special? New evidence from the FDIC-induced failure of healthy banks. American Economic Review, 95,1712–1730.

Beck, T. (2011). Bank Failure Resolution: A Conceptual Framework, in: Panagiotis Delimatsis and Nils Herger (Eds.): Financial Regulation at the Crossroads: Implications for Supervision, Institutional Design and Trade.

Beck, T., Bruno, B. and Carletti, E. (2021). When and how to unwind COVID-support measures to the banking system? prepared for the European Parliament’s Committee on Economic and Monetary Affairs (ECON).

Beck, T., Da-Rocha-Lopes, S., and Silva, A.F. (2021). Sharing the Pain? Credit Supply and Real Effects of Bank Bail-ins. Review of Financial Studies, 34, 1747-1788.

Beck, T., Radev, D., and Schnabel, I. (2020). Bank Resolution Regimes and Systemic Risk, CEPR Discussion Paper 14724

Beck, T., Doettling, R., Lambert, T., and van Dijk, M. (2020). Liqudity Creation, Investment and Growth, CEPR Discussion Paper 14956

Cordella, T., and Yeyati, E.L. (2003). Bank Bailouts: Moral Hazard vs. Value Effect. Journal of Financial Intermediation, 12, 300-330.

DeYoung, R., Kowalik, M., and Reidhill, J. (2013). A theory of failed bank resolution: technological change and political economics. Journal of Financial Stability, 9, 612-627.

Dell’Ariccia, G., and Ratnovski, L. (2019). Bailouts and systemic insurance. Journal of Banking and Finance, 105, 166-177.

Dell’Ariccia, G., Kadyrzhanova, D., Ratnovski, L., and Minoiu, C. (2021). Bank lending in the knowledge economy. Review of Financial Studies, forthcoming.

Eisert, T., and Eufinger, C. (2018). Interbank Networks and Backdoor Bailouts: Benefiting from Other Banks’ Government Guarantees. Management Science, 65, 3673-3693.

Enria, A. (2020). Crisis management for medium-sized banks: the case for a European approach. Keynote speech at the Banca d’Italia workshop on the crisis management framework for banks in the EU.

ESRB (2021). Financial stability implications of support measures to protect the real economy from the COVID-19 pandemic. Frankfurt a.M., Germany
Farhi, E., and Tirole, J. (2012). Collective moral hazard, maturity mismatch, and systemic bailouts. American Economic Review 102, 60-93.

Hardy, B. (2021). Covid-19 bank dividend payout restrictions: effects and trade-offs. BIS Bulletin 38.

Haskel, J., and Westlake, S. (2017). The rise of the intangible economy: Capitalism without capital. Princeton University Press, Princeton.

Keister, T. (2015). Bailouts and financial fragility. The Review of Economic Studies, 83, 704-736.

Laeven, L., and Valencia, F. (2018). Systemic banking crises revisited. IMF Working Paper 18.206.

Langfield, S., and Pagano, M. (2016). Bank bias in Europe: Effects on systemic risk and growth. Economic Policy, 31, 51-106.

Leonello, A. (2018). Government guarantees and the two-way feedback between banking and sovereign debt crises. Journal of Financial Economics, 130, 592-619.

Repullo, R. (2005). Liquidity, Risk Taking, and the Lender of Last Resort. International Journal of Central Banking, 1, 47-80.

Sandbu, M. (2020). The corporate zombies stalking Europe, Financial Times, 8 September.

Single Resolution Board (2019). Public Interest Assessment: SRB Approach, Brussels, Belgium.

Walther, A., and White, L. (2020). Rules versus discretion in bank resolution. Review of Financial Studies, 33, 5594-5629.

World Bank (2016). Bank resolution and bail-in in the EU: selected case studies pre and post BRRD. Working Paper 112265. World Bank.