Surreal Realities of the CDS Markets – Part 1

In 2008, several three letter ‘weapons of mass destruction’ (ABS, CDO, SIV etc) threatened the stability of the financial system. Another – CDS (credit default swaps) – made a cameo appearance especially in September as Lehman Brothers and AIG experienced their ‘fifteen minutes of Warholian fame’.

In 2009, as the global economy slows and the risk of corporate defaults increases, there will be increasing interest in the role and performance of CDS contracts. It remains to be seen whether they emerge as instruments of risk reduction or risk creators. Insurance by Another Name In a typical CDS contract, the buyer of protection transfers the risk of default of a borrower (the reference entity) to a protection seller who for a fee indemnifies the protection buyer against credit losses. The economics of a credit derivative contract are similar to an insurance contract. However, if the CDS is classified legally as an insurance contract, then depending on the jurisdiction there can be problems. Entry into the contract may be illegal unless the seller of protection is licensed as an insurance company. This may result in potential penalties for the seller of protection and render the contract unenforceable at law by one or both parties. The buyer of protection (as the insured party) may be subject to the duty to act with utmost good faith (uberrimae fides) requiring full disclosure of all risks associated with the reference entity. Market practice dictates that the seller must rely on its own inquiries as to the reference entity. In the absence of disclosure, the seller of protection (the insurer) may not be liable to make any payment in the case of a credit event if the CDS was regarded as an insurance contract. The buyer of protection must have an insurable interest in the underlying risk and must be able to identify the specific loss suffered. In practice, this may not always be feasible in a credit derivative contract (especially where the buyer of protection is ‘short’ – it does not have a position in a bond or loan of the reference entity). Lawyers take the view that CDS contracts are not insurance contracts. At least, that’s what I think they are saying in legal opinions but as with anything to do with lawyers there is room for doubt! The reasoning is worthy of a master of rhetoric – as payment is triggered by an event and not by a proven loss and you don’t have to have a loss to make a claim therefore the CDS cannot be an insurance contract. Like a lot of aspects of the CDS market, the reasoning is circular – the outcome is used to prove the premise. As Adlai Stevenson would have put it: “These are the conclusions on which I base my facts.” Evolution & Creationism The buyer of protection is hedging the risk of default of the reference entity (this has similarities to shorting a bond or loan). The seller of protection is assuming the risk of default of the reference entity (this has similarities to buying a bond or loan). This analysis only applies if the CDS contract is held to maturity or until the reference entity suffers a credit event. In practice, CDS contracts may be traded prior to maturity. This allows market participants to use CDS contracts to trade and profit from changes in the fees charged (effectively, the market’s valuation of the risk of default for the reference entity at a given time). For the buyer of protection, the CDS contract avoids the need to transfer loans or bonds to hedge the credit risk of the issuer or borrower. This may be useful for illiquid bonds and especially loans, where it may be difficult to transfer the debt without the consent of the borrower. It allows disaggregation of key elements in hedging credit risk such as timing of the hedge, maturity of the hedge, currency in which the hedge is transacted and the pricing of the hedge. This increases flexibility in hedging credit risk. The documentation for CDS contracts is less expensive and less complicated relative to that needed for selling or transferring a loan. The transfer of the risk of a loan can be completed without disclosure to the ultimate borrower. This is possible as the loan is not sold or transferred but hedged through the separate CDS transaction. For the seller of protection, the CDS contracts allows entities other than traditional financial institutions with lack of credit origination infrastructure to participate relatively easily in the credit market. In essence, anybody can enter the ‘lending’ business. The CDS contract, being off-balance sheet and unfunded, allows a seller of protection to take positions in credit markets on a leveraged basis; that is, without investing the full face value of the loan or bond. The CDS contract facilitates short selling credit risk (similar to short selling bonds). This overcomes structural issues, such as the illiquid nature of the corporate bond repo market, that make it difficult, in practice, to short sell credit risk. The volume of CDS contracts is also unconstrained by the available amount of the reference entity’s outstanding bonds and loans potentially increasing the overall liquidity of credit markets. The CDS contract and the entire structured credit market were predicated originally on hedging of credit risk. Today, CDS volumes are estimated to be roughly three to four times volumes of underlying bonds or loans. This is inconsistent with the use of CDS as a “pure” hedging instrument. Over time, the market has changed focus – in Mae West’s words: “I used to be Snow White, but I drifted.” The ability to short credit, leverage positions, and trade credit unrestricted by the size of the underlying debt market have become the dominant drivers of growth in the market for these instruments. This reflects increased interest amongst investors, such as hedge funds, in trading credit risk. Promises of Performance Where banks have used CDS contracts to hedge credit risk, the key issue is whether the contracts protect the banks from the underlying credit risk being hedged? As Mae West also noted: “An ounce of performance is worth pounds of promises”. Documentation and counterparty risk means that the market may not function as participants and regulators hope if actual defaults occur. CDS documentation is highly standardised to facilitate trading. It generally does not exactly match the terms of the underlying credit risk (for example, the bond or loan) being hedged. A CDS contract is only likely to be a close hedge to another position in an offsetting CDS contract. CDS contracts are technically complex in relation to the identity of the entity being hedged, the events that are covered and how the CDS contract is to be settled. This means that the hedge may not provide, in practice, the protection sought. In fairness, all financial hedges display some degree of mismatch or “basis” risk. Mistaken Identity

For each corporate grouping only one or in some cases a few reference entities are traded in the CDS market. This helps promote trading liquidity. There may be substantial mismatches in “who” is being hedged with “whom”. For example, if there is a guarantee then the effectiveness of the hedge may depend on whether the entity being hedged is an “upstream”, “downstream” or “sidestream” subsidiary to the reference entity. If there are corporate actions (takeovers; mergers; leveraged buyouts), then the reference entity can change according to a set of complex rules. A hedging bank may end up “hedged” on a counterparty that it has no exposure to. A bank seeking exposure to a particular credit can end up with exposure to an entity that it did not intend to take exposure to. Unlike the “real” credit market, sellers of protection in a CDS may not have “a seat at the table” to protect their interests. The problems relating to the identity of the entity being hedged may affect the restructuring of financially distressed firms. For example, in 2007, there were proposals to restructure financially distressed monoline or credit insurers separating the municipal bond insurance business from other activities of the firm. Such a restructure would have affected outstanding CDS contracts. In practice, this, at a minimum, would affect CDS contracts and its effectiveness as a hedge. In the current downturn, it may also potentially impede implementation of corporate reorganisation that may have benefits for policyholders, bondholders and shareholders of the firm. Events & Non-Events The CDS contract is triggered by a “credit event”; broadly, this equates to default by the reference entity. The buyer of protection is not protected against “all” defaults. They are only protected against defaults on a specified set of obligations in certain currencies.

It is possible that there is a loan default but technical difficulties may make it difficult to trigger the CDS hedging that loan. Some credit events like “restructuring” are complex. There are different versions – R (restructuring); NR (no restructuring); MR (modified restructuring); MMR (modified modified restructuring). Different contracts use different versions. The triggering of CDS on Federal National Mortgage Association (“FNMA” or Fannie Mae) and the Federal Home Loan Mortgage Corporation (“FHLMC” or Freddie Mac”) illustrates some of the issues with the credit events in the standard ISDA contract. CDS contracts on Freddie and Fannie were triggered as a result of the “conservatorship”. This may seem odd given the government actions were specifically designed to allow Fannie and Freddie to continue fully honouring their obligations. However, “conservatorship” is specifically included within the definition of “bankruptcy” in the CDS contract resulting in a “technical” triggering of the contracts. This necessitated settlement of around $500 billion in CDS contracts with losses totaling $25 to 40 billion (based on prices established via the auction mechanism described below). The triggering of these contracts poses questions on the effectiveness of CDS contracts in transferring risk of default. “PAI” (publicly available information) must generally be used to trigger the CDS contract. Recent credit events have been straightforward Chapter 11 filings and bankruptcy. For other credit events (failure to pay or restructuring), there may be problems in establishing that the credit event took place. This has a systemic dimension. A CDS protection buyer may have to put the reference entity into bankruptcy or Chapter 11 in order to be able to settle the contract. A study by Henry Hu and Bernard Black (from the University of Texas) concluded that CDS contracts may create incentives for creditors to push troubled companies into bankruptcy [See Henry T.C. Hu and Bernard Black (January 2008) Debt, Equity & Hybrid Decoupling: Governance & Systemic Risk Implications; University of Texas, Law and Economics Working Paper No. 120]. This may exacerbate losses in case of defaults. Getting Physical

In case of default, the protection buyer in CDS must deliver a defaulted bond or loan – the deliverable obligation – to the protection seller in return for receiving the face value of the delivered item (known as physical settlement). Eligible deliverable obligations are defined in the agreement and usually are a sub-set of the outstanding bonds and loans of the reference entity.

Where a reference entity defaults, generally buyers of protection will seek to deliver the “cheapest to deliver” bond or loan to the seller of protection in settlement of the contract. The cheapest to deliver bond or loan is simply the bond or loan eligible for delivery that is trading at the lowest price in the market. This allows buyers of protection to maximise the gain on the CDS contract and conversely increases the loss to the seller of protection. The presence of the cheapest to deliver feature can distort trading in outstanding bonds or loans of a financially distressed entity.

The settlement mechanics may also cause problems even where there is no default. One company refinanced its debt using commercial mortgage backed securities (“CMBS”). The company was downgraded by rating agencies. A shortage of deliverable obligations (the company used the funds from the CMBS to repay its bond and loans) meant that the CDS fee for the company fell sharply (indicative of an improvement in credit quality). This resulted in mark-to-market losses for bemused hedgers. This is known in the trade as an “orphaned CDS”.

What Am I Bid for This Fine Piece of Junk?

On actively traded names CDS volumes are substantially greater than outstanding debt making it difficult to settle contracts. When Delphi defaulted, the volume of CDS outstanding was estimated at $28 billion against $5.2 billion of bonds and loans (not all of which qualified for delivery) [See James Batterman and Eric Rosenthal Special Report: Delphi, Credit Derivatives, and Bond Trading Behavior After a Bankruptcy Filing (28 November 2005); www.fitchratings.com]. In more recent defaults, a similar pattern has been evident.

Shortage of deliverable items and practical restrictions on settling CDS contracts has forced the use of “protocols” – where any two counterparties, by mutual consent, substitute cash settlement for physical delivery. In cash settlement, the seller makes a payment to the buyer of protection. The payment is intended to cover the loss suffered by the protection buyer based on the market price of defaulted bonds established through an “auction” system. The auction was designed by ISDA to be robust and free of the risk of manipulation. In recent years, the protocol auction mechanism was used in a number of high profile defaults. The industry’s view is that the auction mechanism worked fine. In fairness, there were no notable operational issues. However, to argue that the process worked perfectly overstates the success. The following cases highlight some of the issues in respect of the protocol and auction mechanism. In Delphi, the protocol resulted in a settlement price of 63.38% (the market estimate of recovery by the lender). The protection buyer received 36.62% (100% – 63.38%) or US$3.662 million per US$10 million CDS contract. Fitch Ratings assigned a R6 recovery rating to Delphi’s senior unsecured obligation equating to a 0-10% recovery band – far below the price established through the protocol [See James Batterman and Eric Rosenthal Special Report: Delphi, Credit Derivatives, and Bond Trading Behavior After a Bankruptcy Filing (28 November 2005); www.fitchratings.com]. The buyer of protection depending on what was being hedged may have potentially received a payment on its hedge well below its actual losses – effectively it would not have been fully hedged. The auction prices in the settlement of CDS on Fannie and Freddie were as follows:

Fannie Mae – 91.51% for senior debt (the seller of protection would pay 8.49% (100% – 91.51%)) and 99.9% for subordinated debt (the seller of protection would pay 0.01% (100% – 99.99%)).

Freddie Mac – 94% for senior debt (the seller of protection would pay 6.00% (100% – 94%)); and 98% for subordinated debt (the seller of protection would pay 2.00% (100% – 98%)). The lower payout on the subordinated debt is puzzling. Holders of subordinated debt rank behind senior debt holders and would generally be expected to suffer larger losses in bankruptcy. It is likely that subordinated protection buyers suffered in a short squeeze resulting in their contracts expiring virtually worthless. The differences in the payouts between the two entities are also puzzling given the fact that they are both under identical “conservatorship” arrangements and the ultimate risk in both cases is the US government. The settlement prices in CDS Settlements for senior debt in more recent cases were as follows: Lehman Brother: 8.625% (equivalent to payout for the seller of protection of 91.375% (100% -8.625%).

Washington Mutual: 57% (equivalent to payout for the seller of protection of 43% (100% -57%).

Icelandic Banks: 1.25% to 6.625% (equivalent to payout for the seller of protection of 93.375% to 98.75% (100% -6.625% or 100% – 1.25%).

The final payout in Lehman Brothers and the Icelandic banks appears high relative to historical statistics for losses on bank debt. The Washington Mutual payout may have been affected by capital remaining at the holding company, Washington Mutual Inc. (estimated at $2.8 billion). The skewed final prices may not assist confidence in CDS contracts as a mechanism for hedging. ISDA and the industry will shortly ‘hardwire’ the auction mechanism into the standard documentation for CDS contracts. Traders and hedgers will no longer have a choice between physical and cash settlements.

In the case of actual defaults the CDS market may provide significant employment to a whole galaxy of lawyers trying to figure out whether and how the contract should work.  There is the risk that contracts may not always provide buyers of protection with the hedge against loss that they assumed they would receive. It may also exacerbate losses for sellers of protection through technical triggering of the CDS contract (where there is no actual failure) or through the size of the payout. © 2009 Satyajit Das All Rights reserved. Satyajit Das is a risk consultant and author of Traders, Guns & Money: Knowns and Unknowns in the Dazzling World of Derivatives (2006, FT-Prentice Hall).

4 Responses to "Surreal Realities of the CDS Markets – Part 1"

  1. RichardBerlin   January 16, 2009 at 10:49 am

    We need a method to trade dancing electrons (CDS, CDO, etc), which trades are settled based on specific cost and time based amounts of money. This really should not be a problem and I cannot understand what the difficulties are. As there are billions of electrons being traded, and a back of the envelope calculation yields billions of dollars being available, the relation comes to nearly one dollar per electron. Does anyone have another way to calculate these costs?

    • BlackyBlack   January 16, 2009 at 4:02 pm

      O, Richard, don’t you know that the cost of trading electrons cannot be determined until the electrons have stopped trading, at which point their value is either 0 or 1, the costs, when measured in dollars, being not a quantity, but a quality, or, as we say in America, either yes or no.

  2. whiteWhitey   January 16, 2009 at 11:34 pm

    and 0 on one side and 1 on the other ultimately nets to what?a winner and a loser.yes, stop trading. game over.

  3. cdulan   January 23, 2009 at 1:30 pm

    Satyajit,I curious if, in the event of nationalization of Citi, B of A, AIG, or one of the remaining investment banks, that this would be considered a qualified credit event. Although I know it would be the intent of the government to make whole debt holders, is there any risk of a large CDS settlement amount due to the massive amount of contracts involved with these parties.