Wednesday, 27 March 2013

Troika Wrong? Estimates of the Cyprus Debt

Source: IIF publication "Cyprus:Just the Facts"
According to an IIF publication of about a week ago, Cyprus had about 15.6bn of government debt, or about 87% of GDP. Of it, €6.6bn were held by non-residents, leaving about 9bn to residents. Of the €6.6bn of non-resident claims, €3.9bn were held by official creditors (mostly the 2.5bn loan from Russia and another €1.1bn from the EIB and the Council of Europe Development Bank). The rest (or €2.7bn) represent medium- and long-term bonds. It also appears that these are almost exclusively Euro Medium Term Notes (EMTN) issued under British Law.

Of the remaining €8.4bn debt, 1.4bn represent medium- and long-term bonds and 1.9bn represent bank recapitalization bonds for Laiki bank in May 2012. Of the whole €7bn held by domestic creditors one-third or about €2.35bn of them is short-term.

When first reading about the Cypriot bail-out (bail-in would indeed be a better term) I thought the amount would be approximately 10bn for both financing government debt as well as recapitalizing the banks. However, it appears that under the last deal "achieved" by the Eurogroup €10bn are to be given just for government needs and not for recapitalization. 

Now, let us add the numbers presented above: €3.9bn of short-term, non-resident claims in addition to 1.9bn for the Laiki recapitalization and 2.35bn held by residents. This adds up to €8.15bn. Yet, hold on for a second: if we exclude the 1.9bn for Laiki, (which will not be of use since the bank is to be liquidated) then the amount is about €6.25bn. Then, the question becomes on whether the short-term claims could have been rolled-over. Russia has explicitly stated that it would be willing to do so, yet we will not even consider that scenario for now. We will just consider the fact that residents would, which decreases the amount to €3.9bn.

It appears that the Troika is not considering rolling-over loans: it is just thinking about changing the identity of the creditors, from residents to the EU or the IMF. 

Under the Troika scenario of €8.15bn, the Cypriot state has to produce minimal deficits in every year to come, even with private consumption and GDP contracting. The hole is government finances is about 4% or approximately 720m. Under the Troika scenario just two years of the same amount of deficit would force the Cypriots to ask for more money (improving government finances when GDP is shrinking is a midsummer day's dream. Greece can indicate more details on that), thus perpetuating the situation.

Even if Russia would not roll-over its loan (which it probably would), it would meant that the EU would not have to give more than €10bn to Cyprus. As the bank recapitalization costs are now less than 6bn, the whole €17bn package seems way in excess. Why government officials, with every inside information available to them could not see this is beyond my understanding.


  1. Hi Euronomist,
    I would appreciate (not being an economist) if you could clarify what your estimate of an appropriate bailout amount is, based on the above numbers (and assuming the 6bn ceiling on bank recap needs).

    1. Hello Andrew.

      From my point of view, if the bank recapitalization ceiling is 6bn (which again is rather odd given that the Bank of Cyprus needs are about 1.5-2bn but let's take it as our baseline scenario), the whole amount should not be in excess of 10-11bn. This amount includes bank recapitalization, government debt and another 1-2bn for future deficits. Yet, this is with the assumption that the Russian loan will not be restructured. If it is then 8-9bn would be a better price tag (again, including bank recap, debt and a margin of 1-2bn for extra deficits).

      In contrast, the package agreed on the first Eurogroup was of a 19.2bn bailout while the second was of an approximate 16bn bailout, so there is at least a 5bn difference (or about 30% of their calculations).

      The whole depositor mess could have been avoided if they had been more careful with their calculations