Bail out countries but bail in private, short-term creditors
by Daniel Gros (originally published at VoxEU.org)
Despite its large size relative to the small Irish economy, the bailout of Ireland is not working. Risk premiums continue to rise. This column argues that part of the problem lies in a seemingly innocuous provision in the rescue facility that is to replace the current European Financial Stability Facility in 2013. The argument is tricky, but the heart of the problem is the insistence that rescue financing be senior to private debt while simultaneously ruling out rescheduling of short-term debt.
European policy-makers might come to regret the statement of the Eurogroup (Eurogroup 2010) that announced the outline of the new European Stability Mechanism that will replace the temporary EFSF from 2013.
The stated purpose of the announcement was to stabilise financial markets, to provide assurances that funding for a crisis resolution mechanism would be available even after the EFSF ends in 2013.
However, the announcement may very well have the opposite effect. The announcement has already led one rating agency to put Greece’s debt on a negative watch, and it might be behind the further fall in the price of longer-dated Greek government debt since the November announcement.
Indeed, the announcement might end up destabilising markets. The reason is that Eurozone Finance Ministers also announced their intention to make future official financing senior to private creditors. The logic of this is somewhat involved.
Why is seniority for official financing critical to long-term debt prices?
While public-debt seniority sounds reasonable at first sight, thinking ahead a bit reveals problems. Public-debt seniority can lead to the paradoxical situation where large financing packages – like those offered this year in Europe – push up long-term government bond interest rates. The technical appendix goes through the maths of this, but the basic logic can be illustrated with an analogy.
Think of a nation’s creditors as people standing in a queue for bread; short-term creditors are at the front of the queue and long-term creditors at the back – the difference is that in this breadline, everyone has already paid for the bread for which they are queuing. If people fear that the bread will run out before everyone gets theirs, things may turn nasty. This is what is now happening with Greek, Irish, and Portuguese debt.
To steady the situation, the IMF and EFSF drove up with a truckload of bread to convince everyone that this is just a temporary problem. The success of this sort of rescue package depends entirely on which of two realities the troubled nations are facing – insolvency or illiquidity.
If the shop really does have enough bread – the nation really can honour all its debt – such a rescue package can work brilliantly. The sight of piles of fresh bread reassures everyone; things go back to normal. In this version of reality – where the problems stem from illiquidity and mis-guided expectations – the issue of official-assistance seniority matters little. Whether the rescuing truck driver lines up at the head, middle, or back of the queue to get back the bread he brought, everyone gets paid.
If, however, the problem stems from insolvency rather than illiquidity, official seniority really matters. For example, if the bread shop has pre-sold coupons for 100 loaves of breads, but can bake only 90, the average loss among its creditors will be 10%.
If the rescue-package’s truckload of bread ends up fully only in the hands of the short-term creditors at the front of the queue, the eventual 10% shortage will not be evenly spread. The shortage will fall fully on the long-term creditors. For example, imagine that the IMF/EFSF/ESM truck has 50 loaves of breads, which are used to satisfy those in front of the queue.
When the shop finally opens and begins distributing the freshly baked bread, seniority puts the IMF/EFSF/ESM people at the head of the queue; they get their 50 loaves. This means that there will be only 40 loaves left for the remaining 50 creditors at the back of the queue; a loss rate of 20% – twice the average beforehand.
Now imagine that people in the queue are still impatient after the IMF/EFSF/ESM truck has shown up. And in response, the authorities send another truck, with the number plate ECB/SMP/May2010 and this truck brings another 30 loaves of bread (the total rescue package goes up to 80, so 80 out of 100 creditors are fully paid off).
What happens now when the shop opens and the ECB jumps the queue? The 20 creditors remaining in the back of the queue discover that there are only 10 loaves left for them; the loss rate has jumped to 50%.
As many observers believe some Eurozone periphery nations are insolvent rather than illiquid (Eichengreen 2010), this logic can explain the seemingly paradoxical reaction of long-run interest rates – a big rescue package is a signal that losses – if they occur – will be concentrated on long-term private creditors.
How it worked in the first bailouts
Concerns over official debt seniority were already present in the €110 billion package for Greece in early 2010. At that time, the Eurozone donor countries decided that their bilateral credits would not be senior.
This detail of the aid package to Greece was not widely discussed in public although it was not a foregone conclusion given that the bilateral credits came alongside IMF credits, which are generally recognised as senior.
Moreover, in May it was also explicitly decided that the EFSF loans would not be senior because this might reduce the willingness of the private sector to continue financing programme countries. You can read it yourself: It is “FAQ 10” from the website of the EFSF:
A10 – Will the EFSF be a preferred creditor?
No. Unlike the IMF the EFSF will have the same standing as any other sovereign claim on the country (pari passu). Private investors would be reluctant to provide loans to the country concerned if there were too many preferred creditors”
This aspect of the EFSF might have contributed a lot to calming markets over the summer. If nothing else, it is a signal that the official lenders believe it is an illiquidity problem, not an insolvency problem, and are willing to back up this belief by risking their own money.
However, this is now about to change as the Greek and Irish programmes will be transformed into ESM programmes where European rescue funding is senior to private debt (only subordinate to IMF lending). The only rationale for this change given in the 28 November 2010 statement (Eurogroup 2010) is that this is done in order to protect the interest of taxpayers.
But this in itself is a signal to investors. Finance Ministers, which are the ultimate insiders, have effectively said that there is a risk that the bread might run out before everyone was paid and that they do not want to be the ones that come up short.
The signal is quite clear: buyers beware! (Especially long-term bondholders who are at the back of the queue).
What should be done?
The loss-shifting logic discussed above is why ‘bail outs’ of a private company never start by paying back short term creditors. In a typical ‘debtor in possession’ financing, all creditors have to agree to a standstill.
The same principle should be applied to countries. Official financing should not be used to repay outstanding short-term debt at full face value. For example, one of the conditions of IMF/EFS packages could be that the maturity of short-term bonds would be extended. Such a rescheduling would formally represent a ‘credit event’, and would raise interest rates on short-term bonds, but it could lower the interest rates on long-term debt.
Most IMF programmes actually ‘involve’ the private sector in the sense that the IMF usually insists that banks roll over their existing lines of credit. For example, all the recent IMF programmes in Central and Eastern Europe were buttressed by the so called ‘Vienna initiative’ in which the major EU banks were asked to not only to roll over existing lines of credit, but also to provide their local subsidiaries with additional capital. Why should an extension of this principle from bank credit to bonds be regarded as a revolution?
The present policy stance of claiming seniority for official bailout financing runs the risks of triggering a vicious circle. The ever larger resources available for bailouts could be taken as a signal that an ever greater share of eventual losses will be shifted to long-term government bonds. Long-term bond rates would rise. Since these are taken as the key measure of financial market reaction to the package, each subsequent package could lead to higher observed risk premiums which would put even more Eurozone nations into precarious situations that would require yet more, even larger bailout packages.
Moreover, the large scale buying of peripheral debt by the ECB could end up having a similar destabilising effect if the ECB were also insist that it be treated as a senior creditor.
There seem to be only two ways out of this spiral:
- Short-term creditors are bailed in via a rescheduling, or
- The rescue packages are so large that they cover the entire stock of debt outstanding.
The latter would not be compatible with the Treaty and would anyway overtax the fiscal capacity of the core countries. The former would be considered a ‘credit event’ by ratings agencies and would thus trigger most of the negative consequences of a formal default.
Unless European policy-makers rethink their decision about not wanting to take any risk, they have only unpleasant options left.
Eichengreen, Barry (2010). “Ireland’s rescue package: Disaster for Ireland, bad omen for the Eurozone,” VoxEU.org, 3 December.
Eurogroup (2010). “Statement by the Eurogroup”, 28 November 2010.
Technical appendix: An illustrative model
The model has a country with a (foreign) debt equal to one. Assume the future of the country in question is not certain. With probability 1-p the country will be able to repay its creditors in full in the next period (resources available for creditors > 1). In the other state of the world, the resources available for creditors are equal only to q, with (q<1). This implies that with probability p the country will go into default and the sum available for creditors will then be only q which is less than the full face amount of the debt. The variable (1-p) can be thought of as the growth prospects of the country.
The price (value) of the private long-term debt (as a proportion of its face value) will then be given by:
PV = price = (1-p) + pq = 1 – p(1-q) < 1
where (1-q) represents the loss (the hair cut) private creditors have to accept in case of default.
How would the availability of official financing change the equilibrium? The key issue is now whether the proceeds from the official financing are used to retire a corresponding amount of short term debt. This will be the case if there is no ‘involvement’ of private creditors, i.e. no rescheduling of short term debt. This implies that at the end of the operation (which takes place within the current period) the outstanding amount of private sector debt is reduced to 1-s (where s denotes the amount of official financing measured as a share of total debt).
What happens if official creditors are senior to private ones? In the case of default the distribution of the amount available for foreign creditors of the country then becomes the following: the official creditors are fully reimbursed (they receive s) and the remainder goes to the private creditors, i.e. q-s (assuming q>s). The variable s could also stand for the bond purchases of the ECB. In this case the remaining (private) creditors stand to lose out because with its purchase the ECB substitutes a pari passu private creditor.
Keeping in mind that the private debt outstanding is reduced to 1-s, this implies that the value of private debt (as a proportion of its face value) is given by:
Price = (1-p) + p (q-s)/(1-s) = 1 – p(1-q)/(1-s) < 1 (for q>s)
The value of private debt thus falls as the share of official (senior) debt increases. This equation illustrates what common sense would suggest: As s goes towards q (the official sector gets the entire resources available in case the bad state of the world materialises) the value of private debt goes towards 1-p, i.e. the probability of the good outcome under which the private creditors receive full payment. This might explain why the price of Greek and other bonds is so low: the probability of a default might not be very high, but the recovery values for the private sector might be very low.
So what difference would an official financing package make (assuming it is senior)? The difference in the price of (long-term) debt is given by:
Price with official financing – price without official financing = – ps(1-q)/(1-s)
which is clearly decreasing in s. Ceteris paribus, increasing the share of official, senior financing thus lowers the price of debt, or, equivalently increases the risk premium demanded by the private creditors.
Is there a way out of this conundrum? The obvious solution is to ‘bail in’ short-term creditors as is done in any company rescue operation (which always requires a stand still of all creditors). Any package which increases the likelihood of repayment, but does not waste resources on paying creditors whose claims come due in the short run should be a positive for existing long-term bondholders.
The model can also be used to analyze the price of newly issued debt if there is an implicit guarantee on the existing stock of debt. In this case the new debt is effectively junior and its price will be given by:
Price of new debt = (1-p)
This illustrates the difficulties that arise with the often repeated official announcement that all existing bonds will be fully protected. If the stock of old debt is guaranteed the value of the new debt will be low because in case of a default the new debt would be the only one to carry losses. By contrast the price of old debt should be at par (if the guarantee on the old debt is credible).
 Legal scholars will debate the legal value of any assertion that ESM loans will be senior. But, whatever the legal situation it is clear that in reality repaying emergency loans from other member countries will be a more pressing priority for an EU member country than repaying private creditors – especially if they are foreigners.
 This mechanism was explicitly recognised on the website of the EFSF which explains that EFSF financing should not be senior because otherwise private creditors would require a higher risk premium.
 Seniority of official credits would not have posed a legal problem because most (over 95 %) Greek public debt is domestic debt, without a negative pledge clause. The same applies to Ireland and other Eurozone countries. However, it reality it was always difficult to conceive that in case of default Germany (France and other donors) would have accepted a loss on the credits they had given to Greece in the context of the EMF/Eurozone package. Even in the case of Argentina, Spain obtained in the end full payments on the bilateral loans it had given in the context of the IMF packages that had been extended to the country in the vain hope of staving off default.
 The ECB has probably already now larger claims on Greece than what has been disbursed under the 110 billion euro IMF/Euro area member countries rescue package. However, the ECB buys at a discount and in the market. From a legal point of view it might have difficulties insisting on preferred creditor status. However, in case a country has to negotiate with its creditors the ECB would be a very strong negotiating position and might de facto be able to insist of getting at its money back.
 It is debatable whether the ECB could legally enforce any senior creditor status since it bought its holdings on the open market. However, in reality there is very little private creditors could do if a country in default gives the ECB a better deal, which it might have to do in order to preserve access to the ECB’s financing window for its banks. Moreover, the ECB might claim senior status for the collateral it has received from banks (and the banks of a country in default are likely to experience difficulties as well).