Yesterday, Reuters’ blogger Felix Salmon in a well-written if somewhat verbose essay, makes the argument that “Greece has the upper hand” in its ongoing negotiations with the ad hoc and official group of creditors. It would be a great analysis if it wasn’t for one minor detail. It is wrong. And while that in itself is hardly newsworthy, the fact that, as usual, its conclusion is built upon others’ primary research and analysis, including that of the Wall Street Journal, merely reinforces the fact that there is little understanding in the mainstream media of what is actually going on behind the scenes in the Greek negotiations, and thus a comprehension of how prepack (for now) bankruptcy processes operate. Furthermore, since the Greek “case study” will have dramatic implications for not only other instances of sovereign default, many of which are already lining up especially in Europe, but for the sovereign bond market in general, this may be a good time to explain why not only does Greece not have the upper hand, but why an adverse outcome from the 11th hour discussions between the IIF, the ad hoc creditors, Greece, and the Troika, would have monumental consequences for the entire bond market in general.
But before we proceed with the analysis, we should point out one minor nuance: Salmon, and thus the WSJ’s Fidler, are correct that Greece has all the leverage in the world, in the same way that a suicidal person has all the leverage to take their own life as they stand on the ledge of a skyscraper. Because from a strategic standpoint, the reality is that over the past 2 years, the entire financial establishment has done everything in its power to mask the fact that Europe is currently undergoing a stealthy restructuring, without it actually being represented as a restructuring. The reason for this is that while an ex-event of default status quo allows the world’s financial establishment to continue marking sovereign debt, even highly impaired one (remember: central planners are always right, markets – always wrong in pricing risk, or so the central planners say), at whatever prices it desires (recall that one of the very first things to happen in the post-Lehman collapse was the elimination of the Mark-to-Market statute, thus affording banks a plethora of gimmicks to mark ‘assets’ on their books at any valuation that excel spews out based simply on input assumptions, which in some cases are openly fraudulent), a case of sovereign default will very likely make mark to market unavoidable, thus exposing the proverbial nudity of the emperor. It also has implications for the ECB, for CDS triggers, and other consequences, but those are of secondary importance for the time being. Most importantly, the Nash Equilibrium at least until now, had afforded creditors, who in many cases have known very well that they have ‘weak protections’ on their sovereign holdings (more on this in a second), the myth that they are not subject to subordination, or seniority claims on their holdings, and thus the sovereign market was uniform, orpari passu. The outcome of the Greek negotiations, should Greece indeed use the “nuclear option” and force a coercive cramdown on any one, or all, bondholder classes, would do away with this myth in the blink of an eye, and instantaneously create a split between what will hence be perceived as senior and subordinated sovereign bonds. These are all considerations that the ECB, that European banks, and most importantly European sovereigns (and Greece) are all too aware of, and since the need to fund future deficits will only rise, any impairment of the sovereign funding apparatus is not only suicide for Greece, but for Europe, and eventually for the rest of the developed world.
Additionally Salmon ignores a simple tactical observation, one which the hedge funds are all too aware of, namely that while the bulk of Greek bonds are issued under Greek-law (a fact we first observed back in June, when we made the assessment of just who it is that really holds the reins in the default process) and while lacking collective action clauses, can be ‘crammed down’ retroactively, a smaller portion, which is estimated to be between €25 and €40 billion, has been issued under foreign, primarily UK-law, with strong creditor protection, and with Collective Action Clauses, which require that anywhere between 66% and 75% of all creditors agree to a given process, in this case the ongoing Greek prepack exchange offer (more later), for it to occur. It also means that bondholders in all other European countries are carefully watching if contract rights of “strong” UK-indentures are abrogated either in Greece or elsewhere, which would be a signal that there is no sovereign debt in circulation that is safe any longer from future attempts to strip positive and negative covenants, or explicitly stated bondholder rights. This is especially topical, as with Greece about to proceed with a prepack (non) bankruptcy, all eyes will turn to Portugal which is next, and after that Ireland, Spain and Italy. In this regard, what happens in Greece, under the advice of Cleary Gottlieb’s Lee Buccheit, will be seen as a framework for all future bankruptcies that Europe will undergo.
And if Greece does proceed with what Salmon indicates is its “upper hand” course of action, what it will be doing, again going back to game theory, is defecting first, in the process forcing a broad sell off of weak, and potentially, strong indentures bonds of the other PIIGS nations, eventually leading to the collapse of demand for European paper, and the complete loss of confidence in the ECB, which has become a defacto source of equity for the PIIGS, an outcome which will eventually lead to the elimination of all funding for Greece itself. Which is why we said that Greece has as much leverage as one about to commit suicide… but at least it will be first – the line after it long and dignified.
Greek Bankruptcy 101
Before we get into the implications of what a scorched earth strategy by Greece would be, we would like to explain the process as it stands in Greece. Greece, has over the past nearly two years, been the functional equivalent of an insolvent corporation. The hundreds of billions in Troika bailout funding provided so generously to Greece is nothing but a prepetition Debtor in Possession (DIP) loan, with a first lien and collateral protection. The IMF may get paid back, but Greece will say goodbye to half its islands and historical monuments in the fire-sale that precedes. Furthermore, the ECB which as recently estimated by Barclays, has bought about €36 billion of Greek debt, is in effect a provider of equity financing. While this requires a tangential analysis, the ECB does not act as a Greek creditor, whose primary focus is to be repaid. No, the ECB would be more than happy to hold all the Greek debt, as it does not care one bit whether or not it gets paid interest – after all it can just print cash to fund its undercapitalized status should Greek bonds finally be recognized as worthless. If that were the case, Greece would be able to proceed with any debt transaction it desires, as any impairment at the ECB level would be promptly internalized, even if the ECB were to change its charter, which it probably very easily can, to make a Greek event of default a non-event from an accounting standpoint. Yes – the ECB’s credibility will be greatly impaired, but how “credible” was it to begin with? Of course, Germany will hardly be pleased that Draghi is about to foot the bailout of an otherwise insolvent country, and monetize hundreds of billions; yet this would be spun that in doing so, the ECB would be assuring that continuation of the existing way of life… if only modestly longer. In short, this means that the ECB has been acting as a proxy debtor pari passu to Greece, even though it owns Greek debt. Said otherwise, the ECB has been conducting a quiet Greek debt-for-equity exchange, which would have had far greater success if, paradoxically, the market deterioration had persisted after the summer of 2010, when however the Fed proceeded with QE2, and stabilized credit markets (briefly). Of course: Europe can’t devalue its currency alone – Bernanke will not be happy. The point is that if that ECB held all of the Greek sovereign debt, there would be absolutely no difficulties in getting the current “creditor” deal done as the ECB would have agreed to any terms. Especially since the ECB cares not one bit, if its Greek “equity” is impaired all the way to zero.
So we have a DIP lender, and we have continuing debt-for-equty (which has not converted nearly enough debt into equity). What is missing? Why an exchange offer and an actual fresh start balance sheet of course. Which is where the so-far-failing IIF negotiations come into play.
Here the moving pieces are most fluid, and the adversaries are Greek bondholders on one hand, primarily hedge funds who have bought Greek bonds in recent weeks and months and who seek as high a cash payout as is possible, and the IMF on the other, which is trying to make the “fresh start” Greek balance sheet as viable as possible. Because even at a 120% debt/GDP ratio post “reorg” it is hardly a leap of faith to assume that Greece will be insolvent again, and that quite quickly, especially with the country paralyzed by daily strikes, and where the deficit is now well into the double digits. At last check, the negotiations had stalled with private bondholders offered 30 year “post-petition” (said in gest – if Greece gets its agreement, there will be no actual bankruptcy petition, but for all intents and purposes there is) bond with a 4% coupon, which however, the FT just announced, would be cut to 3.5% on IMF demands, making the deal even less palatable for hedge funds. To sweeten the deal, the creditors would also be offered a 15% recovery on par in the form of short-term EFSF bonds, but no actual cash. We leave it up to our readers imagination what happens to the EFSF bond’s price when all the Greek bondholders proceed to dump their allocations at the same time.
Needless to say, this is the stage where the leverage shifts from Greece to the creditors. Because while Greece and the IMF can demand that bondholders suffer 100% losses (even if that means a complete wipe out of Greek pension funds holding Greek bonds – a totally separate topic which we are confident the Greek media will have fun with on its own), exchange offers are never a sure thing, which is why it has never been branded as one for popular consumption, as failure would mean that the creditors have won and that a freefall bankruptcy is imminent.
This is also the stage where the broader media is confused (whether objectively so, or by representing the interests of conflicted hedge funds) as evidenced by the Reuters conclusion. The reality is that in order for an exchange offer to be binding, some majority of bondholders have to agree with the transaction. The problem is that the bulk of Greek bonds do not have what is known as a Collective Action Clause, or a framework which says what percentage of favorable votes is needed to enforce a decision, all have to agree. However, this opens the door for changing the rules. As Citi explained some time ago, “Greek law bonds have no Collective Action Clauses (CACs) which mean that voluntary restructurings require 100% of investors to accept the new terms in order to avoid triggering a default, an almost impossible hurdle.” Which is why the Greek negotiation process implicitly requires the retroactive imposition of CAC, one which on one hand will facilitate the “exchange offer”, yet on the other will create great distrust of any bonds issued under domestic law in other European countries.
Yet where the process falls squarely on its face, is the fact that Greece also has issued a modest amount, somewhere over €25 billion face, in bonds issued under UK-law. These are bonds which already have Collective Action Clauses and which as Stephen J. Choi and Mitu Gulati explain, come in two flavors: “Those that were issued prior to 2004 contained CACs that allow holders of 66% or more of an issue to modify payment terms in a manner that would bind all other holders. The bonds issued after 2004 require the consent of holders of 75% or more of an issue.” Incidentally, this is where the Greece has the upper hand argument fails because while Greece can force local-law bondholders to do pretty much anything, it has no chance of doing that if a given hedge fund cartel has already built up a blocking stake in the UK-bonds. Choi and Gulati go on to state the obvious: “Obtaining approvals from between 66% and 75% of the bonds is likely to be difficult.” And this is where the game gets interesting, because while the bulk of the bonds, or what is now becoming obvious is the junior class, can be impaired with impunity (pardon the pun), it is the UK-law, or the non-domestic indenture, bonds, which are the de facto fulcrum security. And since the notional outstanding here is tiny, it is quite easy to build up a blocking stake in the bonds and to obtain full control of the process, especially since the ECB appears to have been building up its own stake in local-law bonds.
As anyone who has ever overseen or participated in a bankruptcy process, the biggest trump card one can attain is to build up a blocking stake in a fulcrum security (just ask Carl Icahn) . Because it does not matter who has a majority. What matters is who has 33% + 1 of the vote to block any consensual deal. This is what is also known as “nuisance value” because in exchange for their votes, those blocking stake holders can demand anything, and be virtually assured of getting it, in order to allow the restructuring process to continue. This is precisely what the hedge fund hold outs, who started accumulating a block stake in the UK bonds some time in October, figured out in mid- to late-2011. And the fact that the ECB did not, back in 2010 when it was actively buying Greek bonds did not, only made it easier.
In a seminal paper by the IMF’s Manmohan Singh titled “Recovery Rates from Distressed Debt -Empicial Evidence from Chapter 11 Filings, International Litigation, and Recent Sovereign Debt Restructurings”, he does a far better job of explaining the holdout precedent than us:
Under the United States Bankruptcy Code, approval of a plan to reorganize requires the approval of two-thirds of each class of creditors. In response to this requirement,some vulture funds attempt to acquire more than one-third of a company’s subordinated debt, with the object of blocking approval of the plan. By delaying the disbursement of funds to creditors, the holdouts exert pressure on senior unsecured holders to strike a deal rather than suffer further losses of time value of money. As a quid pro quo for their consent to the plan, the holders of a blocking position in the subordinated paper demand a larger percentage recovery than they would be entitled to under absolute priority.
And there is the entire Hedge Fund hold out strategy in a nutshell. Since as we already know the local-law bonds are in effect a junior class to the UK-bonds (and only senior to the ECB’s bonds which are effectively a worthless equity tranche), the bargaining power of the process is now with the one or more hedge funds who control the UK-bond blocking stake. Because while Greece can force the local law bonds to agree to anything, and thus enact a coercive “cram down”, it has no such control over UK-law bonds. At least not explicitly, but more on that in a second.
As it so happens, the bulk of the UK-law bonds have a 2016 maturity as the following Chart from Citigroup shows. In fact, of all vintages, this one is most evenly spread between Greek and UK-law.
So how does the active build up of a blocking stake look like from a pricing standpoint? It looks as follows: in this chart one can easily see the preferential accumulation of a UK-law bond over its less “protected” cousin in recent months as this strategy was being implemented by one or more hedge funds.
As can be seen the price for this preference is as high as 10 cents over the proposed “recovery” value for the entire bondholder class as a whole according to recent IIF leaks. Would one pay 43 cents for a bond unless there was something up their sleeve? Obviously not. Which brings us to a whole new topic of “sovereign litigation arbitrage” (prepare to hear this phrase much more in the future). But before we get there, there is one more open question: is it possible that the ECB does in fact hold the trump card, and can negate a blocking stake in the UK-bonds? Again we turn to Choi and Gulati:
The reason the ECB’s large debt holdings are important to the story is that the power to hold out is limited by the fact that, even in the English?law bonds, there exists a mechanism to quash the holdout. Specifically, a large enough fraction of the holders (between 66% and 75% of the bonds in principal amount) can collectively choose to cram down a restructuring on the holdouts. We do not know precisely what fraction of the various English?law bonds the ECB holds. But presumably it is a non?trivial amount, leading the bondholders who might be contemplating holding out, to be concerned that the ECB might use its votes to force a deal on them.
It may not be trivial, but it certainly is not sufficient. Because when one thinks that of the €25 billion in calculated face non-local bonds out there (of which some are non-UK law), the hold outs have to merely control a third plus one or just under €9 billion. At recent prices this is about €3 billion. A €3 billion investment to control the restructuring of a €240 billion (excluding the Troika’s DIP loan) balance sheet? Not bad at all.
So now that we know more or less what the hedge fund strategy is, what happens if one does in fact assume that Greece has the upper hand, and that it is willing to proceed with terminal game theory defection and blast contract law into smithereens only to get a short-term respite?
First, as a reminder, here is how JP Morgan’s Michael Cembalest described the potential next steps if indeed Europe proceeds with the scorched earth, aka, strip all UK-covenants, process.
Will Greece put “collective action clauses” (CAC) in place? Without getting too detailed, many Greek bonds were issued under language known as “universal consent”, which means that all creditors have to agree to changes to maturity, interest or principal. A CAC allows the issuer to obtain a plurality of support from bondholders for changes to the bond indenture, and then impose them on any holdout creditors. There’s nothing wrong with CACs, except for the fact that applying them retroactively changes the rules of the game, and makes a mockery of the quaint notion of contract law. As we explained in Appendix C in our 2012 Outlook, contract law protections for investors in sovereign debt are very weak. Don’t like retroactive CACs? Go sue in an Athens court; good luck to you.
A couple of points here: Cembalest is correct that those pursuing an Orphan Bond option (more on this later) will likely see an uphill climb. However, it is also true that as Singh points out, some of the best recoveries in all distressed work outs come precisely from orphan bonds. Litigation arbitrage gamesmanship aside, however, the JPM Pvt Wealth CIO has nothing to say about UK-bonds, because if the CACs of those indentures are stripped, and overriden with a new set of CACs, which is explicitly what would need to happen for a Greek pari passu fresh start bond market. then all bets are off, as it would mean that the very premise behind indenture protection is now at the mercy of lawmakers on a case by case basis. And just like MF Global being caught red handed commingling client funds was an event that crushed many investors’ confidence in the stock market, so a strong-indenture cram down would have a comparable effect on the bond market.
Incidentally, here is Singh on Orphan Bonds and why they themselves can be so appetizing to distressed investors:
Distressed debt firms prefer holding illiquid debt to liquid debt since it is cheaper butcarries legal rights identical to those of the relatively more expensive liquid debt. One example of illiquid claim is orphan bonds where the majority of a specific bond has either been extinguished via regular amortization prior to default or, has been given a new CUSIP’ (identity) number following a debt exchange. For example, market sources indicate that Argentine orphan debt was keenly sought after the default and has already been bought by distressed debt accounts. Preliminary data from Bloomberg and market sources indicates that three main denominations of Argentine debt were sought after by distressed debt accounts. These were the 12.125 percent coupon 2019’s, where about $102.5 million remained outstanding from the original $1.43 billion; the 10.25 percent coupon 2030’s, where about $240.5 million remained outstanding from the original $1.25 billion; and the 12 percent coupon 2031’s, where about $15.2 million remained outstanding from the original $1.175 billion. In this example, hold-outs have full payment in mind (including accrued interest) and with double digit coupons, interest arrears could be sizeable as the restructuring will most likely be protracted.
Prominent distressed debt accounts in the United States (WL Ross & Co, Oaktree, Cerberus, Angelo Gordon, or their affiliates) usually look for inexpensive claims, provided opportunities from the U.S. corporate distressed debt market do not “crowd out” investment into junk emerging market debt.
At this point it may be worthwhile to take a detour into…
Collective Action Clauses
While a staple in US corporate bond indentures for a long time, Collective Action Clauses (CACs) are a relatively new development in the sovereign bond market. Elmar Koch explains:
Collective action clauses (CACs) are a new element in the international financial architecture which is to ensure orderly and timely resolution of sovereign default. It was only in the summer of 2002 that a Working Group of the G10 was set up with the explicit aim of providing guidelines or a framework for the formulation of these clauses. The proposal by the Working Group gained wide currency with its endorsement by the G10 Finance Ministers and Governors in September 2002. At the same time, US private sector trade associations (“Gang of Seven”) also developed their own proposals for such clauses and there was IMF support throughout the whole period.
In February 2003 such clauses were for the first time included in a sovereign bond issue under New York (NY) law by a large major borrower, Mexico, and several other sovereign borrowers followed suit during 2003-04. By the beginning of 2004 it had become clear that key elements of CACs, in particular majority action clauses, had been included in this new bond documentation. This feature is expected to contribute to the more orderly resolution of sovereign debt crises by preventing unwarranted creditor holdouts.
Yet ironically, in the case of the Greek exchange offer, it is precisely these bonds that allow some form of plurality to be enforced and to override the government’s attempt to enforce a unilateral decision of creditor stripping.
CACs are an integral part of the bond contract between a sovereign borrower and a private sector lender. These clauses become effective when and if a default of a sovereign borrower occurs. The vast economic literature coping with assessing the debt sustainability of a sovereign borrower is thus relevant. In the international context, a sovereign borrower may default on its bonded debt for a variety of reasons which reflect the ability and willingness to honour its debt obligations. From a legal perspective it is easy to claim pacta sunt servanda (contracts have to be honoured), but from a humanitarian/economic or political perspective a sovereign state may assess any debt payments quite differently. However, the CACs discussion usually assumes that the underlying sovereign debt at stake is deemed to be unsustainable.
From an international perspective it is desirable to aim at a resolution mechanism in debt restructuring that has the attributes of fairness (equity) to all parties and is orderly and timely. It should be noted that CACs are not concerned with the substance of the debt negotiation process itself but are primarily concerned with the process of the settlement of litigation within the legal system. Thus any agreement or settlement procedure (negotiation, mediation or arbitration that parties conclude outside the courts) may also be satisfactory and will not necessarily be covered by CACs. The contractual CACs were aimed at two emerging issues: the distribution of a large number of retail bondholders worldwide on the heels of a large credit appetite by some sovereigns, starting with the 1991-92 boom period, and the associated issue that some creditors will attempt to manipulate the process for their own benefit. More recently the emergence of in-fighting between creditors themselves in order to take a stab at assets of sovereign states first has emerged as a serious threat in upsetting orderly and timely restructuring.
The specifics of UK-law and “strong protections”:
Traditionally, CACs were typically included in sovereign bonds governed by English, Japanese and Luxembourg law. Historically, such bonds issued under US, German, Italian or Swiss law did not include such clauses. The largest market for sovereign bonds is in the US, the State of New York. The adoption of CACs on the NY market was thus the key to providing an internationally acceptable level playing field (see Box 1). While Italy adopted CACs in 2003 under NY law, sovereign bonds issued under German and Swiss legislation last year were without CACs.