Credit derivatives grew from an estimated $3 trillion notional amount with a gross market value of $89 billion in the first quarter of 2003 to an estimated $24.3 trillion notional amount with a gross market value of $725 billion in June 2013.
The most common type of credit derivative is the credit default swap. A credit default swap or option is simply an exchange of a fee in exchange for a payment if a credit default event occurs. Credit default swaps differ from total return swaps in that the investor does not take price risk of the reference asset, only the risk of default. The investor receives a fee from the seller of the default risk. The investor makes no payment unless a credit default event occurs.
The traditional or “plain vanilla” credit default swap is a payment by one party in exchange for a credit default protection payment if a credit default event on a reference asset occurs. The amount of the payment is the difference between the original price of the reference asset and the recovery value of the reference asset. The following schematic shows how the cash flow of this credit derivative transaction work:
If the fee is paid upfront, which may be the case for very short dated structures, the agreement is likely to be called a credit default option. If the fee is paid over time, the agreement is more likely to be called a swap. Unless two counterparties are actually swapping and exchanging the credit default risk of two different credits, I prefer to call the former structure a credit default option. Cash flows paid over time are nothing more than an amortization of an option premium. Because the documentation references ISDA master agreements, however, swap terminology has crept into the market. Since the credit derivatives business at many commercial and investment banks is often run by former interest rate swap staff, the tendency to use swap terminology persists. Therefore, I will most often refer to these transactions as credit default “swaps.”
The credit default premium is usually paid over time. For some very short dated structures, the credit default premium may be paid upfront. Professionals new to this market often ask if the premium should be paid upfront, instead of over time. After all, if the credit defaults, the default protection seller will get no additional premiums.
The credit default option or swap is a contingent option, and not to be confused with an American option. A termination payment is only made if a credit event occurs. If the credit event does not occur, the default protection seller has no obligation. The premium can be thought of as the credit spread an investor demands to take the default risk of a given reference asset. If the investor bought an asset swap, the investor would earn a spread to his funding cost representing the compensation, the premium, the investor would need to take the credit default risk of the reference asset in the asset swap.
For an American option, the premium is paid upfront (or over time, but with the proviso that the total premium is owed, even if exercise occurs before the expiration date). The American option can be exercised any time that it is in the money. The holder of the option does not have to exercise, however, and can wait and hope the option will go further in the money. If the market reverses direction, the American option can again become out-of-the money, and the holder who failed to exercise the option when it was in the money cannot exercise. With a credit default option, once the trigger event has occurred, the holder must exercise and the option stays exercisable.
Default protection can be purchased on a loan, a bond, an index of reference obligations, sovereign risk due to cross border commercial transactions, or even on credit exposure due to a derivative contract such as counterparty credit exposure in a cross currency swap transaction. Credit protection can be linked to an individual credit or to a basket of credits.
At first glance, a credit default swap or option looks structurally simpler than a total return swap. A total return swap is a form of financing, and the total return receiver has both market risk and default risk; a credit default swap is embedded in the structure.
The first key difference is that although the price or premium of a credit default swap or option may increase, it is never actually in-the-money until a credit default event, as defined by the confirm language, has occurred. That seems like a knock-in option or a knock-in swap, which is a type of barrier option. Knock-in options have been around since the 1960’s. When a market price reaches a predetermined strike price, the barrier, the knock-in option comes into existence. But this “knock-in” is not linked to traditional market factors, but rather to either credit default or a credit “event.” If the option “knocks in” then, and only then, is the option in the money. The termination payment is usually not binary or predefined, although it can be if both parties agree. The termination payment is linked to a recovery value or recovery rate for the reference credit or reference credits involved.
The terminology is further complicated by the US market’s use of the word swap to refer to an exchange of one bond for another (usually accompanied by a cash payment to make up for any discrepancy in relative values), and the UK market’s use of the term “switch” for the same transaction. US market practitioners are often mystified when they first hear of “asset swap switches,” an exchange of one asset swap package risk for another asset swap package risk.
A variety of structures have evolved in the credit default swap market, and the risk characteristics are dependent on the structure. For example, one structure goes by a variety of names: digital, binary, all-or-nothing, and the zero-one. This structure poses substantial risk. The investor loses the entire notional amount – not merely coupon and some principal loss – if there is a default event.
Other structures such as the ‘par value minus recovery value’ structure can leave a position of premium bonds partially unhedged or can over hedge a position of bonds trading below par. Exposure management officers evaluating the suitability and appropriateness of such deals must be fully aware of the full exposures implied in these transactions.
The credit swap becomes even more interesting when one realizes that the term “default event” does not even apply to many credit agreements. The event, which triggers a termination payment under the terms of the credit default swap confirmation, is negotiable. The event may be defined as a spread widening, an event in a foreign country that may cause its sovereign debt to decline in price, or just about any event upon which the counterparties can agree and define a price. Even the termination payment is negotiable. It may be preset at a fixed amount, or based on the recovery value of a reference asset, to mention only two structures.
Some credit “default” options, those linked to spread widening, for instance, sound suspiciously like put options which are struck out-of-the-money.
A German bank’s emerging markets trader discovered he had exceeded his Russia position limit, an offence for which he could have been fired. He approached a large American bank and asked to buy credit protection on Russia for one month. To accomplish this, the German bank’s trader paid a high premium for a one month, thirty-five point out-of-the money put on Russian Vanesh. The American bank happily pocketed the fat premium for an option the American bank considered virtually worthless. There was no pricing model, no calculation of recovery values, and no analysis of asset swap spreads. The price the German bank’s trader was willing to pay was an obvious windfall for the American Bank. The need defined the price.
Was this transaction a credit derivative or was this merely a bond option?
The credit derivatives market has adopted over-the-counter versus exchange traded terminology, which makes it difficult to define the size of the credit derivatives market. Virtually any credit related over-the-counter option could be defined as a credit derivative. Generally, however, credit derivative contracts are further distinguished from other over-the-counter options on bonds by the fact that they are negotiated transactions.
If an Investor is purchasing credit default protection, what kind of credit default protection seller is most desirable? If prices were the same, a default protection seller with a triple A credit rating and a 0% correlation with the asset the investor is trying to hedge would be the most desirable. But a default protection seller with these characteristics will probably sell very expensive protection. Therefore, it is beneficial to relax the criteria and find another provider. The investor should be aware that there are unsuitable providers, however.
There are unsuitable applications, too. One must as the right questions before trying to apply a solution. Credit derivatives are sometimes seen as the panacea, the answer to any finance problem, which cannot be solved by conventional market strategies. For example, hedging overrated assets in enormous size with a single counterparty that has taken on too much risk can negate the utility of the credit default swap. When you need the payoff, your protection seller may be out of funds.
Asset swap spreads are independent of the credit quality of the investor. A market asset is swapped to a LIBOR based floating coupon, for instance. The market is indifferent to the credit quality of the investor, who pays cash upfront for the asset swap package. Unlike an asset swap, the premium paid to the “investor”, i.e., the credit default protection seller, is sensitive to the credit quality of the investor. The premium is further sensitive to the correlation between the investor and the reference asset on which one is buying the credit default protection. Depending on the structure, the credit default swap contract may require an uncollateralized payment by the investor, if there is a credit default event.
Suppose you have a choice of buying credit default protection from one of two counterparties to hedge a single A rated reference asset. One counterparty is rated BBB with a 0% correlation with the reference asset. The second counterparty is rated single A, but is 90% correlated with the reference asset. Which counterparty is the better choice?
Counter intuitive as it may seem; it is better to buy credit default protection from an uncorrelated lower rated credit default protection seller than from one that is highly correlated with the reference asset. The joint probability of default between the A rated asset and the BBB counterparty could merit an implied credit rating of AA for the credit default protected asset. The combination of the 90% correlated A rated counterparty and reference asset would probably merit a rating no higher than single A.
Determining default correlation then, would seem key. The more information one has on correlation and credit risk, from whatever the source, the better. In the end, it is differences in interpretation of information, which make a market.
Credit Default Swaps are negotiated transactions, and although there is some agreement on what is a plain vanilla structure, there is very little agreement on anything else. Just about any contract imaginable can be created. The key issues in the credit default swap market revolve around the following parameters:
Disputes among credit derivatives counterparties are legion, and in most cases, completely avoidable with good trade craft. Here are a few examples from the early years of the credit derivatives market:
When Enron declared bankruptcy on December 2, 2001, J.P. Morgan Chase Bank had $965 million in losses from payments due on oil and gas contracts with Enron. J.P. Morgan Chase Bank thought the contracts were hedged with surety bonds. The surety bonds were advance payment bonds that guaranteed Enron’s credit risk on prepaid oil and gas forward delivery contracts.
In 2005, Janet Tavakoli addressed the International Monetary Fund and briefly mentioned the famous Conseco dispute where the issue was the deliverables.
For another example of a dispute between Société Générale and Aon see “Crisis in Credit Derivatives”
In the run up to the financial crisis, the volume of OTC credit derivatives exploded, and credit derivatives were abused by malefactors to amplify risk. The same losing tranche of a CDO appeared many times in multiple deals, since unlike cash tranches, the CDS can transfer the same risk over and over, since the CDS merely references the physical asset’s risk. For example, a Senate investigation revealed Goldman Sachs transferred the risk of a nowhere-to-go-but-down CDO tranche thirty-five times.
Another serious issue was unacknowledged counterparty risk. Monoline insurers lost high credit ratings and imploded. For example, although Goldman Sachs and then Treasury Secretary and former Goldman Sachs CEO Hank Paulson denied it, Goldman Sachs was in serious peril for, among other things, its exposure to bailed-out insurance giant, American International Group (AIG).
Banks learned little from the crisis. In 2012, JPMorgan Chase became the poster child for worst risk management practices after its London CIO unit lost more than $6 billion representing years of non-risk-adjusted profits due to oversized risky positions in credit derivatives. JPMorgan took on huge risks for chump change and then denied it to the press, kept important information from regulators and its own board, and was exposed as having poor risk management practices all the way up to Jamie Dimon, the CEO to whom the unit directly reported.
Source: The above description of credit derivatives is based on Janet Tavakoli’s book: Credit Derivatives & Synthetic Structures: A Guide to Instruments and Applications (Wiley 1998, 2001).