The CDS Market
Until 2008, credit default swap (CDS) markets have grown exponentially since their emergence in the late 1990s but shrunk considerably to an estimated outstanding notional value of $9.6 trillion in 2017. Despite deliberate efforts by the International Swaps and Derivatives Association (ISDA) and market participants on both buy and sell sides, the CDS market shifted from growth (up to $58.2 trillion at its peak at the end of 2007) to a contraction due to changes to the global financial regulatory framework. These changes comprise margin and capital requirements on cleared and non-cleared swaps as well as a ban in the E.U. on short selling using sovereign CDSs. Such regulatory changes have reportedly raised costs and decreased demand for CDSs, although CDSs still provide an essential tool for hedging credit risk in financial markets. With a new shaking in the CDS market, provoked by Blackstone’s credit unit GSO Capital Partners’ most recent deal, the rise in costs so far may not be the end as we will see later in this article.
With respect to the Blackstone-Codere and -Hovnanian case, which we will focus on, we want to look closer at “single-name” CDS. The value of outstanding “single-name” CDS, designed to transfer the risk of default of a specific borrower e.g. a corporate, bank, municipality, or sovereign, has fallen from $33 trillion in 2008 to $5 trillion by the 1st half of 2017, according to data from the Bank for International Settlements.
CDSs are financial instruments that provide insurance against a credit event destroying value in an entity’s (usually a corporation’s) debt. The insurer of the credit event (seller) is paid a premium (usually quarterly) over a fixed time period to provide the insurance. And, the insured (buyer) gets reimbursed for any losses in the value of the entity’s debt, if a credit event occurs over the contract’s life. Usually, in case of default, the payment that the issuer does to the owner is equal to the difference between the par value of the bond and the lowest market price among the outstanding bonds of the same seniority: called “cheapest-to-deliver” option (though, it is held by the buyer and not the seller like in Futures on bonds). CDS contracts can then be freely traded on the over-the-counter markets. The contracts are often traded in a distance of days or weeks, even if for bigger companies and in case of credit deterioration the activity of contracts increases. Their price is influenced by many different factors, but the three main ones are changes in the markets, in the macroeconomic conditions and in the firm fundamentals. Spikes in the VIX are usually associated with increasing CDS prices because investors see companies as riskier. Similarly, a decreasing share price is typically linked to increases in CDS prices, because the instability of the stock might hinder the refinancing options for the company, or because it is motivated by company fundamentals. It is mundane that if the debt position becomes unsustainable, sales decline or expenses increase, the ability to make money of the company is influenced, and the probability of debt repayment decreases. This causes CDS prices to increase. Finally, an important price driver in such an illiquid market is Excess Demand (Number of Bids − Number of Offers).
The Codere Case
In 2013, a trading strategy played by Blackstone on CDS over the Spanish gaming company Codere SA had shaken the markets. In the first half of 2013, GSO Capital Partners (the credit affiliate of Blackstone) started buying bonds and CDS of the company, thus engaging in a basis trade. However, Blackstone convinced the firm to manufacture a credit event. Codere should have missed an interest payment on outstanding bonds, so that it would automatically trigger a default of the company, eventually causing the Credit Default Swaps to deliver payments. It is important to notice that CDS deliver payments on a cheapest-to-deliver basis, i.e. the CDS contract allows its owner to deliver any bond of and get paid the difference between the market price at which it is sold and its par value. Therefore, doing such operation on solid companies would not deliver a profit as bonds would be sold almost at par anyway. On the contrary, in Codere’s case the bond was trading at $54.5 cents on the dollar. Therefore, in the event of a default, the CDS owner would get an immediate payment of $45.5 cents on the dollar. What was the good for Codere in all this? In exchange for the missed interest payment, Blackstone and Canyon Partners provided a $100 million loan at a favourable interest rate to Codere, which was at a high risk of bankruptcy as liquidity was draining. Interestingly enough, the technical default of Codere in the December of 2013 had the effect of stabilising the fall of the price of the stock, mainly thanks to the new funding provided by Blackstone.
The Hovnanian Case
The theme of manufactured credit events has come back on the spotlight in the last months thanks to a similar operation that Blackstone has set up with Hovnanian, a United States-based real estate company. GSO has accumulated a big position in CDS sold by Goldman Sachs and Solus, as well as Hovnanian bonds, and is now asking the company to miss a payment. In exchange for this, GSO would grant a 2040 note at 5% interest rate only. Other firms, among which Credit Suisse, are said to have proposed interest rates around 12% for an 8-year note. It is clear that the conditions proposed by GSO are out of the market and extraordinary as the motives for that are in the CDS trade. This caused the mid-CDS premium on Hovnanian swaps to surge seven-fold since the beginning of November.
As the graph shows, the CDS premium surged without a correlated drop in the prices of the 2% coupon bonds. This is because the market does not expect the firm not to be able to repay the principal, but just to postpone the interest payment to enable the CDS triggering. This made the CDS very expensive to purchase. It is worth noting that the 10.5% coupon bonds are trading even above par. However, the incredible price of swaps cannot be justified by this bond. Indeed, this is where the 2040 note comes into play: the newly issued 5% 2040s are trading around $40 cents on the dollar, so well distant from par value. Since CDS payment is based on the cheapest-to-deliver basis, this would make a big gain for Blackstone (and a big loss for Goldman Sachs and Solus). At the same time, it would be a cheap way for the company to refinance its 2019 obligations.
This operation is even more sophisticated than Codere’s one. Indeed, in the first case Blackstone still had a loan to a very risky counterparty (in the low C-rating area) on its balance sheet, which survived thanks to its injection of liquidity. On the contrary, in this case the company is still junk-rated, but it is not as much in financial distress as the majority of outstanding bonds trade around (or even above) par. Therefore, while in the first case the operation carried big risks also for Blackstone, in the Hovnanian one the risk-reward situation is not proportional. This caused the angry reaction of the parties on the opposite side of the CDS, with the hedge fund Solus suing both GSO and the board of Hovnanian. Moreover, contrary to the Codere situation, in this case there were other institutions ready to provide loans, but at a market-rate; thus, the focus has not been on saving the company but to take value from the transaction. We are going to talk about the legitimacy of the operation from a legal standpoint in the following section.
A Legal Overview
Solus Alternative Asset Management LP brought an action against GSO Capital Partners L.P. and Hovnanian as well as asked for a preliminary injunction. The New York court dismissed the preliminary injunction on the grounds that Solus failed to demonstrate the prospect of irreparable harm in the absence of injunctive relief since the impact the transaction was essentially economic and the pre-transaction positions could be restored by a compensation with an amount of money consistent with the economic harm suffered.
Furthermore, the judge asserted obiter dicta that any proliferation of engineered defaults that did occur could likely be mitigated by actions on the part of ISDA. ISDA, whose membership is comprised of CDS market participants, has a process in place to study and approve modifications to its standard documentation, definitions, and master agreement that could change the definition of a failure to pay event.
Indeed, CDS contracts are governed by standard form documentation published by the International Swaps and Derivatives Association (“ISDA”) and the standard form 2014 ISDA CDS contract definitions include the following definition of a “failure to pay”:
after the expiration of any applicable Grace Period (after the satisfaction of any conditions precedent to the commencement of such Grace Period), the failure by the Reference Entity to make, when and where due, any payments in an aggregate amount of not less than the Payment Requirement under one or more Obligations, in accordance with the terms of such Obligations at the time of such failure.
The court considered that ISDA’s mechanisms were sufficient to confront the threat of cleverly engineered defaults. Indeed, the court found that “[ISDA is not] so powerless to act in an effective way with respect to the effect of intentional defaults on the CDS market, given the numerous proposals to prohibit such engineered defaults, as to require an injunction by this Court to prevent irreparable damage to the CDS marketplace.”
In the Hovnanian case, Hovnanian agreed not to make a required May 1, 2018, interest payment of $1.04 million and, thus, the payment default will just exceed the threshold required to trigger a CDS failure to pay credit event. For North American transactions, the threshold payment requirement to trigger a failure to pay is $1 million. This payment default was small, but of the magnitude necessary to trigger cross-defaults on Hovnanian’s other financial obligations.
Finally, in the event of a credit event, the protection purchaser may request that ISDA’s Determinations Committee (the “DC”), which is composed of 10 sell-side and 5 buy-side firms, determine whether such an event has occurred. After a determination that a credit event has occurred, the DC conducts an auction procedure to determine the final price for settlement of the CDS.
Restructuring in CDS Pricing
The CDS premium, when priced arbitrage-free, i.e. neglecting issuers margins, will incorporate probabilities of credit events that can occur to the reference entity and will equalize the present value of the premium payments with their respective probabilities of occurrence and the present value of the expected payment that is due in case of a credit event. In other words, the CDS premium is that quantity that sets the market value of the CDS contract equal to zero at initiation.
In general, there are three events recognised as CDS triggering credit events for companies, namely: bankruptcy, failure to pay required interest (usually subject to a grace period and materiality threshold) and restructuring of one or more obligation. In this manner, restructuring is defined as:
- A unilateral change to material terms of any obligation.
- The mandatory exchange of obligations for new obligations with different terms due to a deterioration in the creditworthiness of the reference entity.
In the case of Blackstone, the CDS is used less as a hedging tool, and more as a way to make a bet on certain credits. Therefore, the settlement of concern is cash settlement.
Consequently, as long as the probabilities persist, it should not be in the interest of the buyer of a CDS to engineer a technical default that triggers the CDS payment and in return offering refinancing at better conditions to the distressed reference entity at an arbitrage-free price. Doing so would mean that the CDS-manipulating buy side is putting as much capital at risk with their refinancing as they can expect to receive in CDS payment.
Following this, it is of interest which events can bring the probabilities to change and thus the CDS contract out of equilibrium as initiated. Such an event would be restructuring of the existing debt as defined above. Yet the vast majority of CDSs written on investment grade entities recognise restructuring. Hence the buyer cannot act upon a restructuring event in this case since the event itself already triggers the CDS. However, it is common practice not to recognise restructuring for high-yield entities. In this case, it is intuitive that the probabilities of credit event occurrence are different. While restructuring events have a direct effect on the probabilities of credit event occurrence by being a credit event itself in the common case of investment grade bonds, they also have an indirect effect on the probabilities since they change the intensity of the other two credit events (bankruptcy and failure to pay interests) in both cases: investment grade and high-yield bonds. The crucial difference is that in the latter case of high-yield bonds, the CDS is still valid and running.
Up to this time, as already said, restructuring events, their probabilities and expected impacts on the companies’ credit event likelihood are indirectly incorporated in the arbitrage-free CDS premium. But, this incorporation sees negative credit events (worsening of the reference entity’s financials) in favour of the buy side and positive credit events in favour of the sell side. With the option of engineering CDS payments, this relationship does not hold true anymore. Negative credit events are still clearly in favour of the buy side, while positive credit events are likewise in favour of the buy side too now!
The positive credit event gives the buy side an incentive to manipulate the CDS by manufacturing a technical default. The payment they receive, that falsely incorporated positive credit events as favourable for the sell side, is more than they would put at risk when refinancing the reference entity’s existing debt that is in default, as its financial situation improved.
In short: the current arbitrage-free pricing is not correct. It fails to correctly incorporate restructuring events (given that the buy side in the CDS contract is capable of offering attractive refinancing to the entity), and benefits the buy side.
There are various possibilities to bring CDS back to fair pricing by changes in certain factors of the contract, anticipating the actual impact of restructuring. For one thing, it is possible to rise the CDS premium to a higher level that restores the equilibrium between present values of premium payments and expected payments due to default. Another possibility is to redefine technical default in the CDS contract such that it is not as easy to enforce payment.