Cyprus and the Troika have agreed a deal that may now keep Cyprus going until the next bailout.  The key features are:

  • Resolution of Cypriot banks raises €4.2bn 
  • The "total programme money" available to Cyprus will be €10bn
  • Because the money raised from Cypriot banks comes via resolution instead of a tax, the Cypriot Parliament does not vote on the deal
  • The shareholders and bondholders (including senior bondholders) in Laiki (Cyprus Popular Bank) are wiped out entirely
  • Laiki is split into a "good bank" and a "bad bank", with the "good bank" being taken over by Bank of Cyprus.  The ECB will then provide "Emergency Liquidity Assistance" (i.e. virtually unlimited loans) to the Bank of Cyprus
  • Depositors with over €100,000 in Laiki will face a haircut of up to 40% (though the final number may be closer to 30%)
  • Senior bondholders of, and depositors with more than €100,000 in, the Bank of Cyprus will also face a haircut, with the amount to be determined according to the amount required to resolve Laiki and recapitalise the Bank of Cyprus.  In the meantime, these larger depositors in the Bank of Cyprus will have their funds frozen
  • Depositors with less than €100,000 will be untouched – they won't even be "insured" as the monies for resolution will all come from bondholders and larger depositors
  • Cyprus will be subject to draconian capital controls for the foreseeable future

This is a far better arrangement than the botched deposit tax the Cypriots tried to pass in the first iteration of the deal.  In particular, it addresses at least two of the three criticisms I made of the previous deal:

  • Bondholders, including senior bondholders, are included and take losses.  This is a very significant event, as it has not previously happened on any scale in the Eurozone.
  • Depositor participation is restricted to bust banks, and takes the form of a haircut in a bank resolution, rather than a tax.  This is an enormously better way to do things.

The third weakness of the original Cypriot scheme was that the €100,000 deposit insurance threshold was not honoured.  On the face of it, it may appear that that has now been done, and indeed most commentators have said this is the single strongest part of the new deal.  However, that is based on a misconception – and an important one.  Depositors under €100,000 have not been insured in the Cypriot deal.  If they had been insured, then a haircut for depositors of all sizes would have been agreed, with smaller depositors then made whole by the national deposit insurance scheme or (if the insurance scheme were not large enough to cover that sum – which it would not have been) by the government.  That was not what happened in Cyprus.

Instead what happened was that smaller depositors were simply exempted from haircuts under the bank resolution – all funds are to be taken from bondholders and large depositors.  That means that the haircut for large depositors is much larger – probably around twice as large – as it would have been under an insurance scheme.

Perhaps some numbers might clarify what I am saying here.  Suppose that the bank were bust (i.e. had assets less than its liabilities) to the tune of €4bn, and there were €20bn in deposits to get them from (and we'll assume no bonds).  And let's suppose that €10bn were deposits of less than €100,000 each, and €10bn were deposits in excess of €100,000 each.  Then under a deposit insurance scheme what would have happened would have been that all depositors took a haircut of 20%, raising 20% of €20bn which is €4bn.  That €4bn would cover the gap between assets and liabilities.  Then the deposit insurance scheme (or the government) would refund €2bn to the smaller depositors, making them whole.  That €2bn would not come from other creditors of the bank, but from wider Cypriot society or whoever else was backing the smaller depositors insurance scheme (e.g. perhaps the Eurozone).

Instead, what is happening in Cyprus is more like this in our toy example: larger depositors take a 40% haircut and smaller depositors take nothing and there is no insurance called upon.  The consequence is that the haircut for larger depositors is twice the haircut they would have been required to make to meet the resolution needs of the bank.  That can be seen in two ways.  One is that the larger depositors are funding the insurance of the smaller depositors.  The other, more straightforwardly, is that the larger depositors are being made "junior" to the smaller depositors – i.e. the larger depositors would have to be entirely wiped out before the junior depositors (and hence the deposit insurance scheme) would lose a euro cent.

So, to say the above again more concisely: under the new Cypriot deal, larger depositors and bondholders in Cypriot banks are being made junior to smaller depositors, meaning that they take around twice the haircut they would otherwise.

In passing, we should not fail to notice a point I have made repeatedly since 2007: absent deposit insurance, the losses for Cypriot depositors would be of the order of 20%, as is typical for depositors in developed country banks.  It is very hard for a bank to lose so much that depositors would lose more than 20% in a bank collapse.  That depositor losses are so relatively modest is a point that policymakers should note and make the public aware of – most people today imagine that in a bank collapse absent deposit insurance, depositors would lose 80%-90% of their money, mainly because there is so little recent experience of bank collapses with depositor losses in developed countries.

Whilst we are on the topic, note that, even absent force majeure, bondholders and depositors not being insured, as I have argued for, is much less inconceivable than most British commentators appear to believe.  The European Commission, Basel Committee, and British governments all now officially back "bail-in" schemes for bondholders, whilst under the Open Bank Resolution framework in New Zealand, there will be no deposit insurance at all.

Returning to Cyprus, the overturning of the property rights of large depositors and bondholders, relative to smaller depositors, has certain consequences.  First, the Russians are likely to regard this as a form of expropriation of the property of foreigners, similar to what was done in the resolution of the Icelandic banks.  Once countries start expropriating foreign property, it's hard to know when to stop.  It seems pretty plausible that when Cyprus gets into trouble again (and we know from Greece, Spain, Portugal, etc. that it will very probably be "when" not "if") it will seek to re-introduce its deposit tax so as to take funds from other Russian accounts (e.g. privatisation receipts of the Russian government, which international agencies encouraged the Russian government to place outside Russia).

Second, if debt contracts can be overturned under such force majeure in Cyprus, why was the same not done in Ireland, Spain, the UK, and elsewhere in 2008?  Why did Irish and Spanish and British taxpayers have to fund the failed loans of bank bondholders and large depositors in Anglo-Irish, Bankia and RBS?

The combination of the botched depositor tax and the over-sized haircuts now being applied to large depositors means that the Cypriot banks will face as fast a run as depositors are permitted to make upon them, as soon as they "re-open".  The consequence is that Cyprus will now face capital controls for the foreseeable future.  That means that Cyprus is no longer part of the single payments area in the euro – and thus not properly a euro member.  It is even possible that, in some senses, a "Cypriot euro" will depreciate against the "non-Cypriot euro".

Cyprus is yet another object lesson for British policymakers.  The UK is one of a number of European countries – Malta, the UK, Spain, Iceland, Cyprus, Iceland, Luxembourg, Slovenia (I call them the MUSICILS) with very large banking sectors relative to GDP.  The Cypriot collapse could easily already have happend to us, and if our policymakers do not up their game, it could still happen to us yet.