Cleaning up the mess: Bank resolution in a systemic crisis

Daniel Gros, Dirk Schoenmaker

06 June 2012



The diabolic loop between the solvency of the banking system and the sovereign fiscal position is now apparent (Lane 2012).

  • In Greece, it is the insolvency of the government that has sunk the banks;
  • In Spain, the banks are sinking the government.

What is common in both countries is that savers are running away when they see the banks and the sovereign propping each other up. Unless the banks in both Greece and Spain are soon recapitalised, the on-going gradual deposit flight might turn quickly into a classic run, the consequences of which are hard to imagine.

We argue that in the case of Spain’s banks, their dire need of capital can only be provided by a European institution, the European Stability Mechanism (ESM). Likewise, given that the Greek government is in no position to prop up its banks, only the ESM can save the Greek banking system. In both cases the ESM, the ECB, and the national central banks should then take control over the banks they have recapitalised, probably best achieved through a new special purpose vehicle (staffed by experts from European Banking Authority).

For the medium term, the creation of a European Deposit Insurance and Resolution Fund (EDIRF) could make the European banking system more resistant to national shocks and the contagion from the Greek and Spanish cases (see our recent research, Schoenmaker and Gros 2012).1 However, the crisis in both Greece and Spain are threatening the survival of the system today and thus require an immediate solution, before the long-run solution can be made operational. The special purpose vehicle used for immediate intervention in Spain and Greece could later be merged into the future EDIRF.

The Commission’s proposals seem again to be a case of ‘too little too late’. The idea of having some co-insurance among national deposit guarantee systems in case a bank with pan-European activities gets into trouble might be useful. But this assumes that there will be any pan-European banking groups left in the Eurozone. The ‘balkanisation’ of the Eurozone’s banking system is progressing now so rapidly that this might soon no longer be the case. Moreover, the problem today comes from local banks, both in Greece and in Spain (where the internationally active banks do not seem to have been heavily involved in real estate lending). In Ireland, the losses came from banks that were mostly local in their activities.

The trouble with troubled banks

Dealing with troubled banks is a difficult topic as every country is a special case. The general theme that emerges in all cases is that a ‘European approach’ is needed when the sovereign is too weak to stand behind its banks. The details vary from case to case. But the general principle is clear: the deeper the hole, the more risk ‘Europe’ will have to take. This is unavoidable given the vital interest of the entire EU in preventing a wholesale collapse of the banking system in any member country. Moreover, the Eurozone taxpayer has already taken large risks given the huge outstanding credits of the ECB towards banks in these countries.

The general principles that should be applied in all cases are simple (see Schoenmaker and Gros 2012 and Allen et al. 2011):

  • The private sector should be ‘involved’, especially in insolvency cases

Equity holders should be aggressively diluted and debt holders should contribute via haircuts or debt-equity swaps.

  • The least cost principle should be followed.

This says that the resolution authority should choose the resolution method in which the total amount of the expenditures and (contingent) liabilities incurred is kept to a minimum.

  • Swift decision-making is essential.

Procrastination leads only to an accumulation of even higher losses and gives private creditors the time to escape any losses by offloading their claims at the government or the ECB.

  • Finally, resolution requires a change in governance to align the interest of management with those who bear the risk after resolution, namely the public authorities.

Bearing these principles in mind, we propose the following:

1. Spanish banks are recapitalised only after full loss recognition

The Spanish supervisor has clearly failed to recognise the depth of the bust in the local real estate market. This is not surprising. All real estate bubbles develop under the premise that ‘this time is different’, or rather that ‘this country is different’. The balance sheets of all Spanish banks must be revalued at a ‘bust’ scenario for the real estate sector (meaning a further large fall in house prices and higher loss rates on mortgages given the ongoing recession, see Alcidi and Gros 2012).

The toughest decision is then what amount of private sector involvement should be required. There is of course a national legal framework for this in the case of a formal insolvency, but it is usually not respected because some groups of creditors are politically too important. The decision of the extent of private creditor participation (dilution of present shareholders, holders of subordinate debt up to senior bondholders) should best be taken by the ESM itself because this institution will be able to weigh the benefit from having to inject less capital against the potential for a destabilisation of the Eurozone banking system. (By contrast, in the case of Ireland the national government had to bear all the cost of preventing a potential destabilisation of the Eurozone banking system through a haircut on senior bondholders.)

Once this has been done and the ESM is satisfied that the restructured cajas (Spanish savings banks) are sound, they could be immediately admitted to a European deposit guarantee scheme. A decisive intervention of the ESM should thus be sufficient to re-establish confidence in the Spanish banking system and stem the deposit flight which has already reached alarming proportions.

2. Stem the deposit flight from Greece

Here, the banking system was effectively bankrupted by the sovereign. Greek banks held Greek government bonds equivalent (in nominal value) to over 200% of capital. The Greek banking system thus had to be recapitalised in the context of the Private Sector Initiative (PSI) operation. The most straightforward solution would have been nationalisation (followed by re-privatisation once the adjustment program had succeeded in stabilising the economy). However, everybody agreed that the Greek government would constitute the worst of all possible owners of the banking system of the country (even for an interim period). Given the self-imposed restriction that the EFSF could only lend to the government combined this led to a recapitalisation via preference shares, which implies full risk for the European tax payer without any control.

In reality, the Greek banking system has de facto negative equity if one puts their claims on the government on a mark-to-market basis and factors in the losses on the existing loan portfolio which only increase as the recession deepens. The solution must therefore be similar: the ESM (via its special purpose vehicle) should take over the banking system, wipe out existing shareholders and assume full control.

The key question in the case of Greece is, however, how to stem the on-going deposit flight that is prompted by a fear that the country might be forced to leave the euro. It is of course not possible to extend a European deposit guarantee to Greek depositors because then the incentives for the government would be clear: if it reintroduces the drachma and converts only loans into the new currency the cost will be borne by the European Deposit Insurance Fund.

However, doing nothing means that the trickle of withdrawals could soon turn into a fully-fledged run. The best might be therefore to make the (new) Greek government the following offer. The ESM/EDIRF could provide a partial insurance for retail deposits (say up to 10 % of the maximum) provided the government agrees to implement the adjustment programme (and thus qualifies for further financial support). Each year, the government continues to implement the adjustment programme the ceiling on the guarantee could be increased. But the entire guarantee would be forfeited if the government decided to stop implementation and exit the euro. This combination would immediately create the strong constituency in Greece for a real adjustment that has been missing so far. Until now public sector employees and pensioners had an incentive to vote against ‘austerity’ to protect their income. With this partial and contingent deposit guarantee there would immediately be millions of depositors who might vote differently to protect their savings.


Alcidi, C and D Gros (2012), “The Spanish hangover”, CEPS Policy Brief, No. 267, CEPS, Brussels.
Allen, F, T Beck, E Carletti, P Lane, D Schoenmaker, and W Wagner (2011), Cross-Border Banking in Europe: Implications for Financial Stability and Macroeconomic Policies, CEPR Report, London.
Lane, P (2012), “The TSCG and Irish Fiscal Policy: Opening Statement to Oireachtas Sub-Committee on the Intergovernmental Treaty”, 4 April, Irish Parliament, Dublin.
Schoenmaker, D and D Gros (2012), “A European Deposit Insurance and Resolution Fund”, DSF Policy Paper, No. 21, Duisenberg school of finance, Amsterdam.
Schoenmaker, D (2012), “Banking Supervision and Resolution: The European Dimension”, Law and Financial Markets Review, 6:52-60.
Veron, N (2011), Banking Federalism is Key to Eurozone’s Survival, Emerging Markets – G20 Edition, 3 November.

1 See more generally, Veron (2011) and Schoenmaker (2012) on the need for a European banking union.



Topics:  EU policies Europe's nations and regions Global economy

Tags:  Spain, bank run, Greece, Eurozone crisis

Director of the Centre for European Policy Studies, Brussels

Professor of Banking and Finance at the Rotterdam School of Management, Erasmus University Rotterdam; Senior Fellow, Bruegel; and Research Fellow, CEPR