This article appeared in the May 2012 issue of Current Economics with permission of the author.
We published a piece before the Greek election to voice our concern on the situation (see “Greece, be worried, be very worried” May 2, 2012). The result of the election is clearly worse than expected, so that an exit from the Euro Zone can no longer be ruled out. We still believe that an exit is not the central case scenario, as it would be extremely expensive for Greece, but also because it is not in the interests of Europe to force an exit. If there is such a thing as rational behaviour, the outcome should thus be that Greece stays in. On the cost of Greece leaving, we published several pieces last year (see Euro breakup – the consequences, September 6, 2011 and Valuing insanity, September 12, 2011). In our opinion, the arguments we put forward are still valid and the numbers we produced are still applicable. In the following, we look at the other side of the argument: what it would cost Europe to keep Greece in, and what it would cost if Greece leaves.
Our view is that the current debt dynamic in Greece is not sustainable. We have written about this issue in the past and anecdotal evidence suggests that this view is widely shared. The question, however, is how much debt restructuring is needed to put back Greece on a sound path. It is difficult to answer this question for at least two reasons. The first reason is that the debt sustainability simulations rest on a number of hypotheses which include the future growth path, the ability of the government to privatise, the level of interest rate charged, etc… The second reason is more political in nature: there is no clear theoretical guidance on what is a sustainable debt level, for each level of debt it is possible to find a primary surplus that would stabilise the trajectory. The question is then “what primary surplus is acceptable over the medium term by a country?” This is essentially a political question.
The point of this paper is not to analyse the debt dynamic of Greece. We thus propose two “reasonable” scenarios, one with a haircut of one-third of the existing debt, a second with a half haircut.
How much would that one-third or a half haircut cost the European taxpayer? Chart 1 shows a breakdown of current Greek debt. It becomes clear that a further restructuring of the debt would inevitably involve the public sector. On our numbers, indeed, the private sector accounts for only about one-quarter of the Greek debt and any restructuring would be insufficient if it involves only the private sector. We think the next step is a Paris Club solution. In our view, the question is “when?” not “if?”.
Here again, this adds a layer of complexity to our estimate of losses for European taxpayers. Indeed, even if one assumes with us that a one-third haircut is needed, restructuring does not need to apply the same haircut to all participants. Private debt could be more affected, but we believe that the International Monetary Fund (IMF) would argue that it is senior and cannot be restructured. On that basis, the European loans would thus enjoy an intermediary status. It is thus not irrational to expect the haircut applied to European loans to be more or less in line with the average.
The math is simple. Europeans have lent in various forms to Greece: via bilateral loans, and via the European Financial Stability Facility (EFSF) more recently but also via the European Central Bank’s (ECB) Securities Markets Programme (SMP). The total exposure is currently €181.9Bn. A one-third haircut on this debt would thus mean a €60.6Bn loss to the European taxpayer, or 0.5% of Euro Area GDP. A 50% haircut on this debt would thus mean a €91.0Bn loss for the European taxpayer, or 0.7% of Euro Area GDP.
In a scenario where Greece leaves the Euro Area, one would need to make an assumption concerning the depreciation of the currency. We assume that the newly introduced drachma could lose half of its value when introduced. This could seem like a large number, but it is consistent with past experience and we would look on it as conservative. In the case of Argentina for example, the peso went from 1 versus the US$ to 3 in a short period of time. (For more information on this subject, please see “Euro Break-up: The consequences” dated September 6, 2011).
Table 1: Cost for Euro Area of Greek Debt Restructuring
Greek public debt is denominated in euros, so the euro face value of the debt would not change if Greece leaves. But the 50% depreciation of the currency means that Greek GDP in euro would be halved too. As a consequence, the debt to GDP ratio would double. This means that, in order to reduce the ratio by one-third, our original target to make the trajectory “sustainable”, the haircut needed is now two-thirds. This simply doubles the loss for foreign investors. In the case of a ½ haircut, it would have to become a 3/4 haircut.
Hence the estimated cost for the European taxpayer of a one-third haircut is no longer €60.6Bn, but potentially €121.3Bn. The cost for the European taxpayer in case of a 50% haircut is no longer €91.0Bn, but €136.4Bn.
This approach obviously does not take into account the GDP disturbances that would likely follow an exit. It can be argued that the reduction of the debt would not be enough after an exit as the GDP trajectory could become significantly worse. We leave this debate aside in order to limit complications, but we need to keep in mind that the situation could be worse than the simple approach we propose.
The story, however, does not end there. If Greece leaves, €104Bn of debt will appear. Where from? Target 2 imbalances. As long as Greece stays in the Euro Zone, the Target 2 imbalances are a mere accounting line which, when aggregated at the Eurosystem level, cancels out. If Greece leaves, this becomes a genuine liability that has to be repaid. In our view we have to assume that the recovery rate would be zero on Target 2, either because it will not be repaid, or because if it is repaid in part, it would be in hard currencies that would not be available for public debt repayment.
We thus need to add €104Bn to the number above mentioned, the total cost for the European taxpayer of a Greek exit would then be €225.3Bn in the case of a one-third haircut (1.8% of the Euro Area GDP), €240.4Bn in the case of a 50% haircut (1.9% of the Euro Area GDP).
In chart 3, we provide more simulations depending on the extent of drachma depreciation. The numbers consistent with the 50% depreciation are shown; the chart goes from a theoretical 0% (Greece leaves the Euro Area, but the drachma remains at parity versus the euro) to an equally theoretical 100% (the drachma is worth zero, hence Greece is unable to repay any debt).
As an aside, but an important one, the losses for the ECB would come from the €55Bn SMP portfolio, but also from the €104Bn Target 2 imbalances. Assuming that Greece is able to repay only one-third or even less of its debt after exiting, this would easily eat up all the capital of the Eurosystem, which currently stands at €83Bn. On that basis, we believe the ECB and national central banks would have to be recapitalised.
The above approach looks at the accounting argument of an exit. We showed that it is considerably more expensive for Europeans if Greece exits rather than stays in. However, the above arguments miss what we would regard as the main risk: indirect spillover effects to other countries. The mechanism that worries us the most would be the likelihood of bank runs in the periphery. If Greece indeed leaves and the drachma loses half of its value, or more, it would become obvious to depositors in other parts of Europe that their deposits are at risk and we may thus see a bank run as a possible scenario. The two IMF countries, Portugal and Ireland, appear to be the most at risk. In this scenario Spain would also be a possible candidate, in our view, and Italy could also come under pressure. But bank runs are difficult to predict and although we believe the risk is high, it is not certain; depositors might regard Greece as a special case.
Chart 4 compares the size of the issue: the Greek debt held by European authorities, the deposits in the banking sector of different countries and the total size of the balance sheet. It shows that an exit by Greece could make the problem considerably bigger if the banking sector was indeed to be dragged into the crisis and if deposits are to be protected.
Chart 3: Cost With Exit of Greece, Depending on Depreciation
How should a bank run be tackled? We think the immediate answer would be liquidity provision by the ECB. This would technically save the banks. Some adjustment in terms of collateral eligible would certainly be needed, but banks which are victims of the bank run would find the funding they may need at the ECB. This would provide support but it would not solve the problem as a number of banks would need to fund a quarter or more of their balance sheet at the ECB. Although the banks would survive, there would be an obvious impact on the availability of bank credit throughout the economy.
A better solution in our view would be to provide a credible guarantee for deposits. A country cannot provide this kind of guarantee on a credible basis: the size of the deposits is too large for a government to back them. It seems to us that the only option would thus be a European guarantee for deposits. This would probably work and reassure depositors. But this also means taxpayers in one country (for example, Germany or Holland) would be backing bank depositors in another (such as, for example, Portugal or Spain). The idea of a European guarantee scheme was debated by the European Commission (EC) last year, but was not implemented for the above-mentioned moral hazard issues. If the solution is technically feasible, we believe it would be an enormous step towards European integration and risk sharing. We are not sure that such a mechanism could be put in place in the short period of time during which bank runs happen. It would have to be put in place pre-emptively. If that was the case, it would presumably apply to all European countries – hence to Greece. This would be yet another incentive for Greece to stay in as they would not be covered by the guarantee if they exit.
Chart 4: Greek Debt and Bank Balance Sheets, Potential Size of a Bank Run vs a Greek Default
Here is the issue, however: imagine you believe that the Euro Zone will break up. The European guarantee would be, by definition, void if a country exits the Euro Area, but that is precisely when you need that guarantee. So you have a guarantee that can be used only when it is of no value. It disappears as soon as you need it. It is a car insurance that becomes void in case of accident; it is thus by no way sure that even a European guarantee would stop a bank run.
Finally, note that capital control is forbidden in Europe. This European rule was introduced even before the Maastricht treaty, as it was a prerequisite for a common currency. So Greece or any other country cannot rely on that measure. In the case of an exit, the measure would almost certainly be introduced, however. Exit from the Euro Zone implies exit from the EU and therefore could abrogate the free movement of other factors of production, such as the mobility of labour and capital.
Deposit freeze is also a dangerous route, in our view. A full freeze is obviously not workable, as it would mean people have no money to spend. Limit to deposit withdrawals usually acts as a signal that the authorities are trying to stop a bank run, which is a powerful signal to depositors that they need to take their money out. This becomes very counterproductive, as the Argentinean experience showed.
To download free samples of our publications, featuring forecasts and analysis covering G7 and Western European countries, please click below:
For further articles from Current Economics, please see News and Articles.