by Andreas Koutras(re-posted by the Greek Economists for Reform.com blog)
The Greek government is currently preoccupied with the solution to the PSI (Private Sector Involvement) enigma. Bringing the PSI to fruition is a precondition for receiving the next tranche of bailout funds from the EU which is necessary to pay its March bond if it is to avoid a disorderly default.
Back in July 2011 the European Council took the decision to share the burden of the Greek bailout with the private sector (as opposed to the official sector i.e. ECB and Sovereigns). This was not surprising at all. The majority of the private investors are institutional investors with well remunerated professional experts and risk analysts. Their investment decision was taken after considering the risks of non-payment. There are very few innocent retail investors with Greek bonds (around 5% or 10billion in total). Thus, it is only natural for them to bear some of the pain of their decisions. One however, could argue against this line of reasoning; Greece misinform them of the true risks, as it provided wrong debt and deficits statistics and engaged in debt beautification practices for many years (as far as I know, no one has taken Greece to court for this). Incidentally, this is a much stronger argument than the one supporting the ludicrous odious debt case.
In all debt restructurings, the practice of burden sharing and of joint workouts is standard and Greece is no exception. When borrowers find themselves in hard times they negotiate with their creditors to find an acceptable solution. What made the EU council decision rather exotic is that it insisted on a voluntary workout, with the ECB being exempted from any burden sharing. The EU insisted on a “voluntary” workout because it was told (correctly) that any coercive restructuring or workout would be an event of default. And European politicians wanted to avoid a default at any cost. Their insistence had more to do with the stigma of being a failure and their abhorrence of rewarding the free market (CDS speculators) rather than any hard logic or financial rationale. But perhaps the factor that was paramount in the minds of many politicians was the taxpayer funding of the bailout. Greece no longer had the sympathy of the average European taxpayer who was effectively paying for Greece’s bailouts. Spending more taxpayer’s money became politically unacceptable and a solution had to be found with the right connotations. In the words of Commissioner Olie Rehn, “We all know what to do, we just don’t know how to do it and get re-elected”. Thus the PSI was born. The EU council’s decision on the PSI could therefore be reworded as follows:
We invite the private bondholders to voluntarily reduce their wealth for the benefit of the common good (us politicians becoming electable again). If you do not wish to participate we would not coerce you or punish you and by the way the largest holder of Greek bonds the ECB would not take part in this wealth reduction.
Any rational person can easily identify the weakness of this proposal and the difficulties that one would encounter in implementing it. To overcome these problems, the EU council added that money would be allocated to recapitalise the Banks which may suffer losses as a result. In other words, the EU is bribing the Financial Institutions into accepting the haircut. Another by-product of this is that EU politicians increase their control over the Banks through the recapitalization of banks with common shares, and this is a politically acceptable move under the current circumstances.
Birth of PSI
The first PSI agreed on the 21st of July 2011 called for a 21% reduction in the Net Present Value of the Greek Bonds. This was achieved by a bond swap whereby the old Greek bonds would be exchanged with new ones that have longer maturity (up to 30Y) and moderate coupons (also under English law for better investor protection). Only bonds maturing up to 2020 were to participate in the first version of the PSI. The bond’s principal at maturity would be guaranteed by the EFSF though the purchase of a zero coupon by Greece.
The 21% reduction was somewhat arbitrary as it assumed a 9% discount yield for the new bonds. In any case it represented the losses of the investors not the savings of Greece. Most bondholders (mainly banks) wrote down their holdings by 21% to 79% as a result. The main reduction of the debt to GDP for Greece did not come from the maturity extension or the coupon but from the zero coupon guarantee. Here is how it worked: Suppose that Greece borrows today 30billion to buy a zero coupon from the EFSF that would mature in 30 years to repay 100billion of the new Greek bonds. Accounting wise this means that the 30billion today would offset 100billion in 30y and the benefit for Greece would be a 70billion reduction in debt. Unfortunately, as the pool of bonds were restricted to maturities up to 2020 and the coupons on the new Bonds were high (average of 5.5%) this did not represent a viable solution for Greece.
This first version of the PSI was doomed to fail for many reasons. A) ECB with approximately 45billion was to be exempted from the exercise, B) the bond swap was voluntary risking the participation, C) only bonds maturing to 2020 were allowed leaving out 61 billion that exist till 2057, D) Coupons on the new bonds were too high and did not reduce the immediate cashflow needs of Greece significantly.
As the unviability of the scheme was quickly realised the restructuring of the first PSI restructuring thus became a reality Apart from the fact that the savings were not enough the PSI was greatly undermined by the ECB which is allowed to free ride the process and the voluntary manner which permits other bondholder to free ride on the back of the ECB.
Mutation of PSI to PSI+
On the 26th of October 2011 the EU leaders took the head of the IIF (representing the bondholders) Mr Dallara into a room and convinced him to accept a 50% haircut, this time in the nominal value of the bonds. Mr Dallara according to reports protested that this was far too big a haircut to be voluntarily accepted by the bondholders but his objections fell into deaf ears. Following this EU Council’s decision most European financial institutions lowered the value of their Greek bonds to 50%. They thus took a big loss. The PSI+ remedied some of the flaws of the first PSI by including all the bonds issued by the republic of Greece and by insisting on a 50% haircut on the nominal rather than the Net Present Value but did nothing to address the ECB as a free rider or the voluntary manner that allows too many to escape the haircut.
Here is the essence of the enigma that the Greek and IIF negotiators need to unravel. If the haircut is too low then there is no benefit to Greece. If on the other hand it is too high (like 50% or more) then bondholders will hold out as the process is voluntary. The financial odds are in their favour. Firstly the process is voluntary and the ECB is not taking part. They can easily free ride with the ECB. Secondly, they have already marked their bonds at 50% and thus they have taken the loss. It is a sunk cost. If they do not participate they could get 100 back. If on the other hand they are somehow coerced into the PSI+ they would get roughly 50 back, namely a limited loss. Faced with this bondholder rebellion the Greek side appears to have panicked and started talking of introducing Collective Action Clauses (CAC) to force the hold outs into participating.
Before entering the fascinating world of CAC’s let me explain the reasons the official sector ( i.e. the ECB ) refuses to take part in the voluntary exercise. The European Central Bank has bought around 45-50billion (nominal amount) of Greek bond in the futile effort to restore stability in the Greek bond market back in 2010. The price it has paid for these Greek bonds is not 100. Market estimates the average purchase price paid by the ECB to be around 75-85% of par. If the ECB marks them down to 50% because of the PSI+, it would represent a loss of 13billion or more. Against this loss the ECB has equity capital of 10.8billion. This means that the ECB would have to raise capital from its shareholders to cover the losses. The shareholders of the ECB are the National Central Banks (NCB) of the Eurozone. This would be a huge blow to the reputation of the central bank and it would also mean taxpayer’s money mainly from the major northern European countries. It has therefore ruled out participating in the voluntary exercise. Similarly, if Greece were to default then all banks including the ECB would have to write down their holdings to account for the default, meaning insolvency not just for the Greek banks but also for the ECB who is there to support the Greek banks! The easy solution of printing money to cover losses is not within the mandate of the ECB and it would directly contravene the Treaties which prohibit the ECB from bailing out countries of the Eurozone.
Collective Action Clauses
Collective Action Clauses (CAC’s henceforth) or Majority Action Clause have a long and varied history. Their introduction served and serves a very real purpose and need. That of facilitating the restructuring or the amendment of the (T)erms & (C)onditions of a Bond by the majority of Noteholders and to avoid rogue, recalcitrant or malevolent holders of bonds holding at ransom both the debtor and other creditors. What are CAC’s? What purpose do they serve? Why does Greece want to introduce them? Let me start with an example of how CAC’s might work.
Suppose that ACME an issuer of debt, finds itself in financial difficulty. The management of ACME and many of the holders of its debt agree that with a bit of help and reorganization ACME can survive and thus repay what it owes. All it needs is to restructure its debt by moving some coupons and extending the maturity by a few years. Although this is the majority view since the bonds do not have a Collective Action Clause (CAC) this majority cannot prevail. ACME needs the consent of 100% of the bondholders. Thus a malevolent or simply a bondholder with a grudge can stall the reorganization of ACME’s debt and therefore cause its demise unless he is paid in full. For this reason the issuers of debt introduced CAC that allow for the majority (or supermajority) view to prevail over a minority interests.
As always, however, there is another side to the ACME story. It could be that the real villain is ACME and the victim is the unsuspecting small bondholder. ACME purposefully, introduced CAC’s with a low majority threshold, say 50% (simple majority) with a view to make non-payment or restructuring fast and easy. ACME colluded with some bondholders to haircut the bond’s principal by 50%. Then ACME found some other way to compensate (recapitalize or sweeten) its friendly creditors.
Thus, CAC’s have two sides (see Greece on CAC Warpath, Devil’s Advocate) . Striking a balance between the real need to restructure a company and with the moral hazard it introduces is a fine art. Ultimately, any dispute in the motives would have to be resolved by courts. In fact, there have been cases where the majority has been overturned as it was not deemed to serve the interests of all bondholders. This means that even though CAC’s allow for the majority view to prevail, the minority bondholders cannot be ignored completely. The majority must act in good faith when imposing a solution. Playing devil’s advocate in the case of Greece we have the official sector not participating, the Greek banks being compensated for their losses (issuer state recapitalizing them) and similarly for other European Financial institutions. We thus have a situation, whereby, certain classes of investors (Official Sector ECB, Greek Banks etc) are indifferent to the restructuring losses. This is by no means a just and equitable burden sharing. It looks as if their vote has been bought by the issuer.
Moreover, one can argue that there is collusion between some investors and the issuer to defraud certain small class of investors by invoking the rights of the majority. If the introduction of the CAC by the Hellenic Republic is done at the minimum of 50% (plus 1) rather than a supermajority of 75% or more that view would be reinforced.
As we have argued above, some investors may recoup their losses through recapitalizations or other means while others take a big hit. This is grounds of discrimination of investors. Additionally, the official sector (ECB) may be exempted from any CAC’s effectively subordinating every other bondholder to the Official sector which is again an event of Default. Finally, punishing the investors is not an acceptable goal for the activation of CAC.
One argument against the introduction of CAC’s has to do with price and liquidity. Studies have shown that the introduction of CAC’s in less creditworthy countries (lower rating) have an adverse effect on the price and liquidity. Namely, creditors demand a higher yield as they view the CAC’s from the moral hazard point of view. For borrowers of high rating CAC’s are viewed as a safety feature.
Europe and the Greek Case
One has to note that while in the case of corporates all countries have laws dealing with bankruptcy (Chapter 11 in US etc) the equivalent for sovereigns does not exist. There have been various attempts by IMF and others to formalize the process with various degrees of success. Introducing CAC’s is one way to ease and facilitate the speedy resolution of debt restructurings. It is for this reason the European Finance Committee (EFC) in Europe welcome in April 2003 the implementation of CAC in European Bond issuance. It may not come as a surprise that Greece never implemented CAC’s in its domestic bonds. In fact, Greece is the only country alongside Italy that it did NOT implement CAC’s in their domestic bond issuance. Italy’s formal claim was that it would adversely affect the price and liquidity of its bonds. We do not know the reasons of the Greek non-compliance but if they are the same with those of Italy, i.e. that it would raise the borrowing cost of Greece, then the Hellenic Republic would be insincere and opportunistic in its arguments that the retrofitting of CAC would not be a credit event as it does not impair the value of the bonds.
Retroactive introduction of CAC’s
Greece has threatened to introduce CAC’s into the Greek bonds by a simple legislative parliamentary process. Greece has further argued that doing so would not be an event of default. There are many issues that are raised if Greece decides to go down this path which throw serious doubts to the non-default claims of the Hellenic Republic (See Greece on CAC Warpath)
In order for an issuer to change unilaterally the terms of a bond contract and simultaneously to avoid default, the change must not impair the value of the bond. But as we have seen, countries like Italy specifically excluded the introduction of CAC’s for exactly this reason. Academic studies have shown that the introduction of CAC’s into a Bond’s prospectus results in a higher yield demanded by investors especially for lower quality issuers such as Greece (See Eichengreen & Mody 2000). Furthermore, the future rights of the bondholders are adversely affected just by introducing them. For these reasons the introduction of the CAC impairs the bonds and as such ISDA may consider this to be an event of default.
- Even if ISDA rules against the default in the previous case, the forced restructuring is definitely an event of default according to the ISDA rules. Any restructuring that is binding to bondholders is a Credit event. See section 4.7 of their document.
- What would happen to the ECB? CAC’s are binding to all bondholders. Including the ECB means that ECB would vote for its bankruptcy (Yes vote) or destroys the PSI (NO vote). Excluding the ECB would be challenged in the courts of law. It would further mean discrimination between bondholders and would effectively subordinate everyone to the official sector (again an event of default). This would cause havoc in the markets. Italy would be the next in line as it does not have CAC either. An unintended consequence of the exclusion of the ECB from the CAC’s is that now it would be easier to block the PSI. Why is that? Because now a bond-holder with only 15% of an issue might jump to 25% or more if the ECB holding is taken out of the voting. In other words, it can play either way. It is simply unpredictable and immensely messy.
- Whatever happens, Greece would have to deal with numerous legal challenges for many years to come. A lawyer’s paradise. In the end, chances are that Greece would have to compromise in order to clean up and re-join the capital markets.
Therefore the introduction of CAC to force a higher participation in the PSI process may cause more trouble than good. It may cause default which would be detrimental not only to Greece but it would affect Europe, the ECB and the functioning of the European debt markets with unintended consequences.
This is the enigma. Finding a way to increase the participation and thus maximize the benefit for Greece, while keeping the voluntary character in order to avoid the default of Greece and the bankruptcy of the ECB.
A possible solution to the Enigma
The restructuring of the Greek debt should aim at:
- Exit Yield.The yield that Greek bonds would be trading after the restructuring
- The exit rating. The rating of Greek debt after the restructuring.
- Sustainability of the Greek finances after the restructuring
In its current form the PSI+ does not fulfill these three essential aims. Either because the bondholders demand high coupons or there are too many holdouts risking the voluntary participation rate. Solving the enigma or at least improving the probability of success may actually be easier than one thinks. Here is how it can be done within the framework of the voluntary PSI (see also Proposal in English and in Greek) and without altering the EU council decisions. Simply:
- Add the voluntary option for the ECB to sell its holdings at purchase cost (75-85%) to the Issuer.
- Add the voluntary option for the private bondholders to exchange their bonds for a cash payment of around 30% in addition to the option of receiving a 30Y new Greek bond.
The word voluntary is stressed in order to keep with the requirements of the PSI and also to avoid price discrimination. Greece does not make a tender offer to buy the bonds back from the ECB. The bond market is an Over The Counter (OTC) market and prices are privately negotiated. Given the volume size of the ECB holdings, the illiquidity in the market and the needs of the buyer and the seller, prices could vary and are negotiable. This does not represent a stealth recapitalization of the ECB. The capital of the ECB was never reduced as a result of its purchases within the Securities Markets Program (SMP).
Assuming that the ECB voluntarily sells back its holdings at cost we immediately change the odds of the free riding. Bonds under Greek law (the majority 93.1%) offer very little protection to the bond holder. They do not have CAC, Negative Pledge, or Cross Default and Acceleration or other protecting clauses in them. Hence, there is very little incentive for holdouts to pursue their strategy once the ECB is taken out. Currently, most free riders are holding out because the ECB is refusing to participate in the PSI restructuring process. By cashing the ECB out we reduce significantly the incentives of free riding. The focus of the free riders or holdouts would very swiftly move to the non-domestic law bonds. These are the objects of desire for holdouts. They offer much better value and protection for holding out. Only hard core investors and possibly retail would remain. We expect around 10-15% of the private bondholders to stay out. This is not an unreasonable proportion as Greek bonds under non-domestic law (mainly English and Swiss) represent 6.9% of the total debt. It would be safe to assume that not all of these bonds are in the hands of holdout investors. The retail proportion is estimated at 5% so an overall 15% of holdouts could also be conservative. Now that the ECB is out of the way the rest of the bondholders would be faced with the choice of participating in the PSI process or not. If they do not, then there are two possibilities. Either they get repaid in full or they risk being restructured with much worse terms than those who participated. This is because the new Greek bonds would probably have features that make them more appealing (English law, Cross default etc) and safer for investors. Holding the new bonds would be more preferable to holding the old ones. Any future restructuring would be harsher to the old bonds than to the new ones. Moreover, the Hellenic Republic could de-list the old bonds after the Bond swap making them very illiquid and also less appealing. The risks also increase if they hold long dated bonds as the cumulative probability of restructuring increases.
In order to increase the participation and maximize the benefit for Greece, the current proposal also adds a cash voluntary sell-back option that would run concurrently with the bond swap option. The purpose of adding the cash offer to the private bondholders is to entice many to sell back their bonds to the Hellenic Republic rather than enter into new 30Y Greek risk. Now that the chances of a successful free riding are reduced to the foreign law bonds and to the nearest maturities and possibly to some hard core investors, Greece can try to maximize the benefits. Let us run some numbers to see how this might work. Greece would pay 36billion to take out the ECB at 80%. Participation in the cash offer would cost a further 54billion (85% participation at a price of 30%). Hold outs would cost another 32billion (maximum and not immediately payable). Thus with around 122billion Greece wipes out roughly 260billion of bonds (rough indicative numbers, see full details on how here and Q&A here). Adding the rest of the debt that Greece has in the form of loans, T-bills, or unstructured indebtedness brings the Debt to GDP to less than 120% instantly. No need to wait for 2020. The money to implement the proposal would again come from the same source as before (Troika loans and/or EFSF or its successor). A further bonus for Greece comes from the fact that the Troika loans have a 10Y grace period and are at a 3.5% interest. This is a significant saving for Greece and it improves the cashflow sustainability.
In fact, right from the first version of the PSI a portion of cash (20billion) was assigned for buybacks and the current PSI+ bond swap offer includes a 15% cash-upfront and a 30Y Greek bond with possibly guaranteed principal at maturity (from the EFSF or other). The current proposal improves on these. The reason the 30% price has become an acceptable alternative is because most of the Greek bonds are now trading between 20-30%. Offering an amount close to 30% would be at a premium to where many are trading. It is important to stress that this cash option was in the minds of policy makers right from the start. Many suggested a kind of reverse Dutch auction last year to buy back and destroy the bonds. The crucial difference is that now it can be implemented with a high probability of success due to the huge drop in prices. Most long dated bonds trade in the range of 20-30%.
The strategy outlined above has many merits for all parties involved. For example:
Advantages for the Hellenic Republic:
- Greece could clean up the majority of its bond stock. Greek debt to GDP would drop to less than 120% from day one. Not in 2020.
- Reducing the outstanding debt improves significantly the chances of Greece being upgraded very fast. Thus the aim of a high exit rating becomes a reality.
- Not having many bonds trading would push the exit interest rate down for Greece, possibly to less than the currently assumed of 9%.
- As the loans taken have a 10Y grace period the sustainability of Greek finances is improved. The 3.5% paid to the Troika loans is lower than the one demanded by investors in the new bonds. This would make further NPV gains for Greece
- The risk of activating some of the moratorium clauses present in the Hellenic Railway bonds and of causing a Credit event is minimized.
- The Republic avoids the introduction of English law in new bonds (only official loans) or at least keeps it to a minimum. Also minimizes the risk of lengthy and costly litigations.
- Greece would most probably be placed on Selective default until an agreement is reached. This is the same status that Greece would be placed if the PSI goes ahead.
- A firm backstop is introduced for 10 years. It would give the necessary time for the much needed economic reforms to work.
- Most importantly it would reverse the extremely bad investment environment and halt the depression both in the economy and in society.
Advantages for Bondholders:
One may ask, why bondholders should prefer the cash offer compared to the new 30Y Greek bonds. There is a variety of reasons. For those who cannot see the attractiveness we list the reasons below:
- Most European banks have marked their holdings close to 50% after the EU council decisions some even lower. They have already taken the hit. In addition, they are asked to take a further NPV loss of possibly 10% with the new Greek Bonds making the total loss close to 60%. Under this light the 30% cash looks like good value.
- They also risk being restructured again if Greece’s finance deteriorates further or if the PSI does not work. Accepting the cash resolves this problem.
- If they accept the 30Y Greek bond they would take back Greek credit risk for 30 years on top of the interest rate risk. A holder of Greek bonds that was exposed to a 5Y Greek credit is now forced to accept 30Y one. This is not very appealing to either a trader or the risk manager of a bank.
- There is no need for Regulatory charges or other provisions (Basle III) due to the 30Y new Greek bonds. And no special accounting provisions
In the end, some may choose to swap their Greek holdings for the new 30Y Greek ones, but financial logic would weigh heavily in favor of the cash offer rather than new 30Y Greek risk.
Advantages for Europe:
- The proposal operates within the framework agreed by the EU council. i.e. Voluntary restructuring and avoiding a hard default. There is no need for a new decision at least with regards to the main points. Only the total amount would need to be updated.
- It further punishes the reckless investors much harsher as the private investor’s haircut is 70% (or more if agreed). The total effective haircut for the Greek bonds would be around 55%.
- ECB does not go bankrupt because of the GGB holdings. The sellback price can be structured so as to limit the losses.
- The voluntary character of the solutions allows the possibility of differential pricing.
- Europe does no need to worry about the financing needs of Greece for 10Y
- Europe implements a market solution for the debt rather than one dictated by political will which has caused so much consternation.
- There is more than a good chance that the retrofitting of CAC or other coercive measures could be avoided. This would ease the pressure of contagion and pressure towards other markets (Italian etc).
- Europe avoids the costly consequences of a Greek default. Whether an orderly or a disorderly default occurs, Europe would pay a much higher price to keep the EU united.
- Europe still has leveraged on Greece through the Stability Pact and the funding of the Greek banks.
- Europe can also demand strict conditionality for the funding of the sellback to proceed.
Funding the proposal
It is clear that Greece does not have the full funds for the proposal. It is also clear however, that it does not have the funds to pay coupons in the new Bonds should the PSI proceed in its current form. Trying to raise it immediately from taxes would exasperate the deep economic recession. Under the current scheme Greece would get 30billion to implement the guarantees needed for the new Greek bonds and a further 89billion once the PSI has proceeded. The current proposal would require some flexibility in the funds committed in order to succeed but would result in much greater benefit for all parties involved.
In order to proceed with the current proposal, 122billion (maximum amount) would need to be found. However, they do not need to be found or raised immediately. Once bondholder agree to swap their holdings for the cash offer, the EFSF (or its successor) could issue short term T-bills of 1M, 3M, 6M, 9M, 12M to exchange the Greek bonds with. Doing so, would ease the pressure to raise all the cash immediately and would spread the cost and effort over a year or more. In this way, bondholders are also compensated for the waiting time.
The current form of the PSI poses many challenges that stem from the irrational character of some of its demands and structures. The voluntary character coupled with the refusal of the ECB to participate in the process make the PSI an unsolvable enigma for the Greek and IIF negotiators to solve. Here we show how with some simple modifications of the PSI scheme, one can significantly increase the chances of the PSI succeeding. This is achieved by removing the largest free rider, the ECB. The additional cash offer (haircut of 70%) to the private bondholders is done in order to destroy much of the outstanding bonds and thus improve the exit rating and the exit interest rate of Greece. Doing so would give Greece a fighting chance for the years ahead.
 See, Klein & Juhle, Majority Rules: Non Cash Bids and the reorganization sale. American Bankruptcy Law Journal, Vol.84, p.297-326.
 ECFIN/CEFCPE(2004)REP/50483 final
 In 2002 the G10 working group adopted this and issued model clauses
 Eichengreen & Mody, Would Collective Action Clauses Raise Borrowing Costs? An Update and Additional Results, http://escholarship.org/uc/item/46p4z4c4
 (a) “Restructuring” means that, with respect to one or more Obligations and in relation to an aggregate amount of not less than the Default Requirement, any one or more of the following events occurs in a form that binds all holders of such Obligation, is agreed between the Reference Entity or a Governmental Authority and a sufficient number of holders of such Obligation to bind all holders of the Obligation or is announced (or otherwise decreed) by a Reference Entity or a Governmental Authority in a form that binds all holders of such Obligation, and such event is not expressly provided for under the terms of such Obligation in effect as of the later of (i) the Credit Event Backstop Date and (ii) the date as of which such Obligation is issued or incurred………