11 min read 25 Feb 10
Summary: The market in sovereign credit default swaps has sprung to life over the past year as worries about the health of nations, rather than corporates, have multiplied. Problems in Dubai in December, and Greece right now, on top of a general deterioration of the developed economies’ budgetary positions have seen sovereign CDS making headlines. Here’s a chart of some of the latest CDS spreads on the major economies showing that fear of sovereign defaults is growing. This article is in the form of a Q&A and aims to answer some of the main questions that our clients are asking us about the sovereign CDS market.
A: Credit Default Swaps (CDS) are contracts made by two market participants to either increase or reduce credit exposure to an entity – in this case a sovereign nation rather than a company. Quoted in basis points per year, a CDS price indicates the cost per year to either buy or sell exposure to the possibility of a sovereign defaulting or restructuring. Selling protection means you receive the premium every year of the contract but bear the risk of capital losses in the event of default; buying protection means that you pay the premium but will receive a payment equivalent to the losses suffered by bond holders in the event of default or restructuring. In other words sovereign CDS behave a little like insurance contracts – you can take the role of the insurer, or be insured.
A: You can use CDS to hedge an existing government bond position against losses it might suffer as sovereign credit worthiness deteriorates, or to take exposure to sovereign risk and receive yield in exchange for that credit risk. Like all derivatives they can be used to hedge trading positions and efficiently manage portfolios, or to take speculative or naked positions on the underlying markets. Because the instruments are mark to market, profits and losses from movements in the CDS levels occur constantly – you do not need to see an event of default to either make or lose money, just a movement in the market’s perception of the risk of that default. These are OTC (over the counter) derivatives, so participants need to have legal documentation in place with counterparties, and collateral moves from one counterparty to the other to reflect movements in the profit and loss – this reduces counterparty risk.
A: The Sovereign CDS contract is triggered when a credit event occurs. There are three credit events for sovereign CDS, and they differ from corporate CDS in that “bankruptcy” is not one of them (that’s a concept that only legally applies to a corporation). The credit events are:
Sometimes credit events are not clear-cut. For example, what about the UK’s War Loan example that we talked about recently? Bond investors voted overwhelmingly to accept the reduced coupon payments (from 5% to 3.5%) – but was there really an alternative? So although this was a voluntary restructuring, nowadays this would probably trigger a CDS credit event. A committee of market participants sometimes has to decide whether such an event has taken place or not – obviously this is not as “clean” as you might like, and there is obviously potential for conflicts of interest to occur.
A: The investor who has bought protection through a CDS contract needs to receive a payment equivalent to the face value of the contract that they have entered into, less any recovery on the bond. So, let’s assume that Country X misses a bond coupon, and its bonds fall to 20 cents in the dollar. There are two ways of settling the contract – physically or simply as an exchange of cash. In physical settlement, I give the counterparty who sold me protection $100 nominal value of bonds for every $100 which I have insured. The counterparty gives me $100 in exchange. Thus if I own the bonds and have hedged them with CDS then although my bonds have fallen by 80%, I get my full face value back. If I have simply been speculating, then I can buy the distressed bonds in the market for 20, and sell them for 100 to my counterparty. Contracts will typically cash settle however – in this case there is an auction to set an observed market price for the distressed bonds and thus determine their recovery value. Instead of bonds changing hands, the counterpart simply transfers the difference in value between the nominal value of the bonds and their distressed price to the buyer of protection. This is neater, but there are again some issues about the fairness of the price determined at auction if participants were conflicted – in a large, liquid government bond market this is less likely to be a problem than in an auction for illiquid high yield issues.
A: No. The contract determines which bonds are deliverable – only bonds in major currencies are allowed, and bonds must be under 30 years to maturity. The concept of the cheapest-to-deliver (CTD) bond is important. The buyer of protection will want to deliver the lowest cash priced bond to the seller of protection, in order to maximise their payment. All other things being equal, this would be a long dated, low coupon bond – although in a full blown default where no coupons are being paid all bonds will tend to trade at similar cash prices in any case. Other issues to consider include whether agency bonds are deliverable into the CDS contract – some believe that KfW bonds are deliverable into a German sovereign CDS contract, while Railtrack bonds are not deliverable into a UK contract, even though both are AAA rated and government guaranteed. Lawyers love credit default swaps.
A: There has only been one example of the auction process being used to determine the recovery rate for sovereign CDS. After the last Ecuadorian default, the auction settled at a recovery rate of 31.4%. In recent sovereign defaults in the days before CDS, the observed recovery rates ranged from 6% (Russia in 1998) to 90% (Dominican Republic in 1990). The average according to Credit Suisse was 39%. Long term recovery rates may of course differ from those used in the auction process. Recoveries will tend to be based on willingness to settle, rather than asset coverage. In a corporate default, debt investors can seize physical assets or contracts from a company; with the exception of overseas embassies and property it is much more difficult to seize sovereign assets.
A: Yes, with the caveat that like all financial instruments, prices are driven by fear and greed and may not reflect the fundamentals. You need to have an assumption for a recovery rate – let’s use 39% as the average. So if Greek 5 year CDS is trading at 400 bps per year, this means that in any one year you anticipate a pre-default spread of (4% x 100/(100-39)) = 6.56%, which given markets are efficient (!) must equate with the expected one year default rate. So on the back of an envelope, ignoring the impact of compounding and the expected timing of a default, the cumulative expected default rate for Greece over the next 5 years is 5 x 6.56%, or over 30%.
On today’s CDS levels, the following default probabilities can be calculated:
|5 year CDS, bps||5 year implied cumulative default at 39% recovery|
A: It wouldn’t be a good idea, and luckily it’s not allowed. UK banks don’t trade UK sovereign CDS (and US banks similarly don’t trade US CDS). If you trade UK sovereign CDS the most liquid contract is in US dollars; you can’t trade the contract in £ – this is because in the event of a sovereign default there is also likely to be a currency crisis which would significantly distort the recovery value in non-£ currencies. Because the UK only has £ obligations (there is no foreign currency debt) the recovery value will therefore be determined by auction and there will be cash settlement. This distortion of the knock-on currency devaluation is the big difference between sovereign CDS and corporate CDS (there is no currency impact in a typical corporate bond default), and a reason why the two instruments cannot be directly compared (ie. you can’t automatically conclude necessarily that because an index of corporate CDS is trading at a lower spread than the UK sovereign then the corporate credit risk is lower).
A: The basis is the difference between the spread over the risk free rate on a bond issued by a government and corporate and the CDS spread. As a simple example, Greek 5 year bonds yield 440 bps over similar maturity German government bonds (risk free, hopefully), and the CDS is at 400 bps. This difference of 40 bps is the basis – it’s common for the CDS level to be below the bond spread (a negative basis). The difference should be arbitragable, but issues such as liquidity, different investor preferences and uncertainty over the cheapest to deliver mean that the differences can persist. It is worth mentioning that given most assets contain an element of sovereign risk, the desire to hedge this out, especially amongst banks, has had a huge technical impact on what is still a fairly illiquid market thus distorting prices. It is not unknown in the credit markets for CDS spreads and corporate bond spreads to move in opposite directions – this can be a real pain trade for those who thought they had hedged an exposure.
A: Some people think so, and there have been rumours in the past few days that the European Commission is considering a ban of some sort as speculators have been blamed for the current Greek woes. Certainly CDS provide a method of reflecting bearish views on a nation in a way that was difficult to do historically (most investors cannot physically short government bonds), and it’s a method that produces a highly visible bellwether of a country’s perceived risk. But Greece, and the other weak sovereigns, would be in trouble even if the CDS market did not exist – the problem is one of fiscal imprudence and lack of will to resolve this, and the market reflects the fundamentals rather than drives them. There is a wider, and perhaps more valid issue with CDS though, and that’s the issue of moral hazard. I can’t buy insurance on somebody else’s house that pays out if it burns down – if I did it might produce perverse incentives for me, and even if I didn’t turn into an arsonist, it might make me think twice about calling 999 if I saw smoke. Does buying protection on a company or a nation create perverse incentives to hinder a possible recovery for that entity? Is it morally right to hope for bankruptcy, or default, when it might mean hardship for employees, or indeed a whole nation? There’s a debate on just this topic on the Economist website.
The value of investments will fluctuate, which will cause prices to fall as well as rise and you may not get back the original amount you invested. Past performance is not a guide to future performance.