Banking union: the problem of untried systems

Abstract

This article explores the problems the EU and the SRB face in trying to implement a credible system for resolving banks without the use of taxpayer funds as a key part of banking union that avoids the doom loop between indebted banks and indebted sovereigns. It finds that without clear examples of how the system works in practice it is very difficult to provide convincing evidence of what will happen given the large number of options for bailing in, the continuing predilection for bailing out in some states and the lack of fiscal backstop for general threats to financial stability.

Keywords:

• Banking union
• bail in
• bail out
• single resolution board
• too big to fail

Notes

1. See RBNZ (2012) for an assessment of the costs of the different forms of resolution.

2. Contingent convertible bonds that can be written down or converted into equity if the bank’s capital or viability passes below a prescribed trigger value.

3. Clearly, the authorities were well prepared and the SRB and the Spanish resolution authority FROB (Fondo de Reestructuración Ordenada Bancaria) would have sorted out the way forward with Santander before the resolution was triggered.

4. Particularly in the ability to act early while the shareholders are still running the company.

5. In other words that they do not frustrate the process say by refusing to allow a subsidiary to be sold.

6. It is worth noting that the SRF is not simply a source of funds that can be used by the SRB in a resolution as some sort of tax on the banking industry. SRR Art 76(3) makes it clear that “The Fund shall not be used directly to absorb the losses of an entity … or to recapitalise such an entity”. While the fund cannot be used until a bail in of at least 8% of total liabilities has been made, it can only be used for six specific purposes thereafter (SRR art 76(1)) which include guarantees, loans, purchasing assets, contributing to a bridge bank/asset management vehicle, paying compensation to creditors/shareholders and making contributions in lieu if a creditor group has been excluded from the bail in. An expectation that the SRF would be used up to its full capacity, if needed, as soon as the 8% bail in was achieved would help explain some of the opposition to the fund and its mutualisation.

7. Single Resolution Regulation (SRR) – Regulation (EU) No 806/2014 of the European Parliament and of the Council of 15 July 2014 establishing uniform rules and a uniform procedure for the resolution of credit institutions and certain investment firms in the framework of a Single Resolution Mechanism and a Single Resolution Fund and amending Regulation (EU) No 1093/2010.

8. In each case this is to be agreed in collaboration with the relevant national authorities.

9. Singh (2016) explains the convoluted relationship between the two types of plan and the authorities involved. FDIC (2014) illustrates the difficulty in providing plausible plans.

10. If there were a separate macro-prudential agency then this would be included as well and of course if some of the other functions are performed by the same agency then the number of participants would be reduced.

11. In such a case, the Commission would undertake an “in-depth investigation” (SRR art 19(3)) which could delay the process considerably, as is illustrated by the delays in agreeing the support package for BMPS.

12. Usually, the authorities can manage to allow a troubled bank to limp along so that it fails on a Friday evening.

13. It is very difficult to get accurate valuations in a crisis. Many institutions will find they need more liquidity and will hence sell rather illiquid assets at low prices, thereby depressing the market price, and any attempt to raise new capital will be expensive in such an uncertain world. Thus, not only are current prices poor indicators of underlying values in those circumstances but the actions of each institution in trying to solve its own problems makes the prices more unfavourable for all the others in a downward spiral of difficulty (de Grauwe, 2013). All this makes it more attractive for a government bailout as then the government will be able to buy bank assets cheaply and hope to sell them at a profit later, as in the Swedish financial crisis in the early 1990s (Mayes 2016). As Poulsen and Andreasen (2011) record, in the case of Amagerbanken, the main bail in attempted thus far, the initial valuation of the assets and losses was too pessimistic and a smaller bail in was eventually agreed.

14. A governmental wish to do something other than apply the BRRD directly is readily possible as the BMPS example illustrates.

15. These are called the “competent” authorities in the wording of the BRRD, to distinguish them from the resolution authorities.

16. Thus, the FDIC starts when a bank changes from being “well capitalised”, which entails having a 2% margin over the minimum required capital adequacy ratio, to being “adequately capitalised”, which means that is merely above the minimum requirement. (https://www.fdic.gov/regulations/safety/manual/section2-1.pdf, at 2.18-9.).

17. The objectives are: “(a) to ensure the continuity of critical functions; (b) to avoid adverse effects on financial stability … ; (c) to protect public funds by minimising reliance on extraordinary public financial support; (d) to protect depositors [covered by the relevant Directives]; (e) to protect client funds and client assets” (SRR art 14(2).

18. Elliott (2012) points out that this contagion not only applies to holders of similar bailinable debt in other banks but also to the holders of more senior instruments in the same bank as they may fear that the losses will be so great that they will also be called upon to share in the loss.

19. A related worry is that the SRF will run out of money. Although studies of the likely demands on the SRF from failures in a major bank suggest it will have sufficient cover (Schoenmaker 2010; Hüser et al. 2017), they are single bank studies and do not allow for the contagion that may mean that other banks may need to draw on the fund as a whole. While Hüser et al. (2017) allow for direct contagion, in the sense that losses to counterparties lead them to fail as well, they do not consider the indirect contagion caused by a fall in confidence in the market as a whole. The fact that TARP was $700bn, while the SRF will be €55bn suggests that major funding may still be needed from governments. However, the impact on the SRF alone might be affordable under the SRR rules (Huertas and Nieto 2014) with the balance falling elsewhere. Schoenmaker (2010) estimates a possible financing cost for the failure of HSBC of €100bn but then most of the cost falls on the UK not the SRF.

20. Gros (2013) suggests that past experience indicates that actually Germany is the most likely country to invoke systemic stability as a means for intervening early with public money as there is no example in recent years of where any creditor however junior has been allowed to make any losses in the event of bank failure in Germany.

21. Which is the expectation in the case of Banco Popular (SRB 2017).

22. Hüser et al. (2017) explore the possible knock on effect of simulated bailing in for the 26 largest banks in the euro area on financial sector creditors. They find that there are no knock on failures but they do not consider the problems of falling asset prices and a general loss of confidence.