VoxEU Column EU institutions Financial Markets Macroeconomic policy

Deposit insurance after Iceland and Cyprus

Depositors in Eurozone banks are facing a steep learning curve on just exactly what deposit insurance means. This column points out that the precedents set in Cyprus and Iceland show that deposit insurance is only a legal commitment for small bank failures. In systemic crises, these are more political than legal commitments, so the solvency of the insuring government matters. A Eurozone-wide deposit-insurance scheme would change this.
This reposted column corrects an error, due to the editor, that was in the first posting.

The facts are now well known. The largest banks in Cyprus are insolvent, but too big for the government of Cyprus to save – at least if it wanted to avoid the ‘double drowning’ fate of Ireland. The government, trying to rescue banks, found itself needing a rescue.

No surprise, then, that the message to Nicosia was: find €5.8 billion without increasing the country’s indebtedness. This was delivered to President Nicos Anastasiades on Friday 15 March 2013 by an ECB executive board member. The €5.8 billion figure was what Cyprus would have to add to the €10 billion offered by the European Stability Mechanism (ESM).

The ECB threat that drove the deal

The threat was clear and credible. Laiki, the second largest bank in Cyprus, had been living off ‘emergency’ ECB loans (i.e. Emergency Liquidity Assistance loans). Without a rescue package that recapitalised the banks, the ECB would cut these off. The likely result of this would have been the fall of Laiki and of the largest Cypriot bank – the Bank of Cyprus. With these in ruins, financial disarray would have reigned. Depositors’ savings would have been wiped out, and borrowing from the Eurosystem would have disappeared.

As a euro-based banking system doesn’t work without euros, Cyprus would probably have no choice but to exit and to issue its own currency. The ramifications of this have been often discussed (see Eichengreen 2007, Buiter 2012, Buiter and Rahbari, 2011a, 2011b, 2012).

  • This new currency would immediately plummet in value, taking with it the value of Cypriot assets.
  • Euro-denominated foreign liabilities under foreign law could not be redenominated without triggering a default. Without such a redenomination, default would also be inevitable.
  • The Cypriot standard of living would be devastated.

Desperate to avert such a catastrophic scenario, Anastasiades looked to the only source of readily available cash – Cypriot bank deposits. Possibly fearing capital flight and bank runs if he taxed solely the large (and thus uninsured) deposits and also possibly believing that many of the large deposits had been broken into small deposits, he proposed a levy on all bank deposits in Cyprus. Small-deposit holders would be taxed at a rate of 6.75%. Fortunately a proposal that spares small depositors and instead wipes out both the junior and senior bondholders of Laiki, and imposes huge losses on its large-deposit holders, has since replaced this scheme.

Deposit protection

Even though it wasn’t adopted, the extraordinary proposal that small depositors should lose a part of their savings – a proposal that had the approval of the Eurogroup, ECB and IMF policymakers – raises the question:

  • Is there any credible protection for small-bank depositors in Europe? And if the answer is no, should there be?

Members of the European Economic Area – this covers the EU plus Switzerland, Norway and Iceland – are required to set up deposit-insurance schemes that cover most depositors up to 100,000 euros. The directive allows member states to choose among different types of schemes, and the idea is that they will be funded by taxing resident banks.

  • The intent of the legislation appears to be to ensure schemes funded by credit institutions are in place to protect small depositors in the event of the failure of a small or medium-sized bank.
  • The question of how depositors are to be compensated if such a scheme is set up and the scale of a banking crisis is such that the scheme lacks the necessary funds is left unanswered.

Either the directive is not meant to cover such a scenario or it is a classic example of mandating without funding.

The Iceland precedent

Cyprus is not the first time deposits security came into doubt. When the Icelandic bank Icesave went down, the relevant court ruled that the Icelandic government was not legally obligated to repay UK and Dutch depositors in a timely fashion (EFTA Court 2013).1 Here is the key:

  • The court accepted Iceland’s argument that the EU directive was never meant to deal with the collapse of an entire banking system.

It noted that the provision of a scheme that was both financed solely by credit institutions and that was able to guarantee coverage in the case of a systemic collapse would itself undermine the stability of the financial system (para 158) and that the provision of a scheme backed by a legal obligation of the state would have a negative effect on competition (para 164).2

Lessons: Deposit insurance as a political vs legal commitment

The Cyprus and Iceland lesson is clear:

  • Deposit insurance is only insurance for small crises.
  • Sufficiently large bank failures or system collapses are a different matter.

In such cases, small depositors are only safe if the sovereign has both the ability and willingness to compensate them.

Consider the ex-ante ability of Cyprus and Iceland to assure their bank deposit:

  • In 2011, assets of commercial banks in Cyprus with Cypriot parents were about five times Cyprus’s GDP.
  • Including subsidiaries of foreign banks domiciled in Cyprus (and covered by Cyprus’s deposit insurance) the ratio is about seven (IMF 2011).
  • In Iceland, the assets of the three large Icelandic banks were about 11 times as large as Icelandic GDP prior to their collapse.

Looking at this, and noting the precedents set in Iceland, it seems clear that these deposits were not insured for the case of systemic failure. 

Cyprus and bankruptcy realities

When a business fails, its creditors must take their losses, otherwise no market economy would function effectively. If the creditors of banks do not take losses when banks fail then banks will take on too much risk. There are two reasons why it may be desirable to protect small depositors, however.

  • The ‘widows and orphans’ argument that small and financially unsophisticated investors should be protected; and
  • The efficiency argument that It is inefficient for large numbers of small-deposit holders to devote resources toward assessing the health of complex financial institutions.

(Although by now, even the most gullible should be wary of tempting interest rates offered by large banks inhabiting tiny island nations.) Although these arguments are appealing, some might assert that the 100,000 euros of coverage is excessively large.

Avoiding bank runs

A third argument that is sometimes made is that deposit insurance prevents contagion. That is, if depositors at failed banks are forced to take losses then there may be runs on solvent banks as well. The idea is that if each depositor believes that all other depositors will run and as a consequence an otherwise solvent bank will fail, then it is optimal for each depositor to run as well. As a result the bank fails. Observing depositors lose their money at a failed bank is the sort of event that might coordinate the beliefs of depositors of solvent banks in such a fashion.

But the argument misses an important point. There are other ways to prevent self-fulfilling bank runs. Runs on solvent banks can be prevented or their effects mitigated by having a central bank such as the Eurosystem that is a credible lender of last resort. This argument alone provides no need for transfers from either taxpayers or other credit institutions to any of the creditors of a failed bank.


If the EEA is wants to provide credible deposit-insurance schemes that cover the collapse of large banks or (not too large) banking systems, then it needs schemes that are jointly and severally funded. However, it is unreasonable to expect the area as a whole to bail out a particular country’s banks unless it can also supervise that country’s banks. This is problematic for the EEA or even the EU, but it may be possible – at least in the Eurozone – when and if the single supervisory mechanism comes into being.


Buiter, Willem (2012),“Greece and the Eurozone: Political leaders should get off their high horses”, Vox Talks, VoxEU.org, 20 February.

Buiter, Willem and Anne Sibert (2008), "The Icelandic Banking Crisis and What to Do about It", April, reprinted as CEPR Policy Insight No. 26, 29 October.

Buiter, Willem and Ebrahim Rahbari (2011a), "A Greek Exit from the Eurozone: A Disaster for Greece, a Crisis for the World", Citi Economics, Global Economics View, 13 September.

Buiter, Willem and Ebrahim Rahbari (2011b), "The future of the Eurozone: fiscal union, break-up or blundering towards a 'you break it you own it Europe'", Citi Economics, Global Economics View, 9 September.

Buiter, Willem and Ebrahim Rahbari (2012), "Rising Risks of Greek Eurozone Exit", Citi Economics, Global Economics View, 6 February.

Buiter, Willem, Ebrahim Rahbari, Juergen Michels and Guillaume Menuet (2012), "EMU Crisis Outlook: Lender of Last Resort on the Way", Citi Economics, Global Economics View, 28.

Eichengreen, Barry (2010), “The euro: Love it or leave it?” VoxEU.org, 17 November 2007; reposted 4 May 2010.

EFTA Court (2013), "Judgement of the Court (on deposit-guarantee schemes – Obligation of result – Emanation of the State – Discrimination)", Luxembourg, 28 January.

European Union (2009), "Directive 2009/14/EC of the European Parliament and of the Council", Official Journal of the European Union, 11 March.

IMF (2011), "Cyprus: 2011 Article IV Consultation", Washington, DC, November.

Pirker, Benedict (2013), "Case E-16/11 ESA/Iceland: It might be called a life jacket but it doesn’t mean it's built for emergencies", European Law Blog, 30 January.

1 EFTA Court (2013). Iceland was bound by an earlier version of the directive – Directive 94/19/EC, which had less stringent wording. The revision suggests that policymakers realise the shortcomings of the directive.

2 It is possible that the Court of Justice of the EU would have a different interpretation. The same question can be put to both courts with different interpretations. See Pirker (2013).

5,039 Reads