PDF | Credit derivatives were at the centre of the recent meltdowns in the financial sector. The article analyzes The article proposes regulatory prescription of 'minimum board responsibilities' and mandatory risk assessment. . 'sub-investment grade' component showed an is about streamlining the trade in derivatives. Credit Derivatives: Trading, Investing and Risk Management. Файл формата pdf; размером 3,08 МБ. Добавлен пользователем. portfolio. An investment bank can use credit derivatives to manage the risks it incurs credit derivatives market to continue its rapid growth, market System, tvnovellas.info
|Language:||English, Spanish, Arabic|
|ePub File Size:||22.60 MB|
|PDF File Size:||11.58 MB|
|Distribution:||Free* [*Sign up for free]|
Credit Derivatives Risk Management, Trading & InvestingGeoff Chaplin MA, DPhil, FFA Credit Derivatives For other. Credit Derivatives: Trading, Investing and Risk Management, Second The credit derivatives industry has come under close scrutiny over the. The book is accompanied by a website which contains tools for credit derivatives valuation and risk management, illustrating the models used in the book and.
The credit derivatives industry has come under close scrutiny over the past few years, with the recent financial crisis highlighting the instability of a number of credit structures and throwing the industry into turmoil. What has been made clear by recent events is the necessity for a thorough understanding of credit derivatives by all parties involved in a transaction, especially traders, structurers, quants and investors. Fully revised and updated to take in to account the new products, markets and risk requirements post financial crisis, Credit Derivatives: Trading, Investing and Risk Management, Second Edition, covers the subject from a real world perspective, tackling issues such as liquidity, poor data, and credit spreads, to the latest innovations in portfolio products, hedging and risk management techniques. The book concentrates on practical issues and develops an understanding of the products through applications and detailed analysis of the risks and alternative means of trading. The book is thoroughly updated to reflect the changes the industry has seen over the past 5 years, notably with an analysis of the lead up and causes of the credit crisis. The book is accompanied by a website which contains tools for credit derivatives valuation and risk management, illustrating the models used in the book and also providing a valuation toolkit.
He has also published many articles in Risk, the Journal of the Institute and Faculty of Actuaries , and others, speaks regularly at conferences and is the author of Credit Derivatives: Pricing and Risk Management: Request permission to reuse content from this site.
Undetected country. NO YES. Credit Derivatives: Trading, Investing, and Risk Management, 2nd Edition. Selected type: Added to Your Shopping Cart. The credit derivatives industry has come under close scrutiny over the past few years, with the recent financial crisis highlighting the instability of a number of credit structures and throwing the industry into turmoil.
This is intended to cover administrative errors and omissions, and other events which might make payment impossible in the very short term. Brazilian debt in US dollars. The claim amount is a key element in the valuation of credit derivatives including bonds so we shall introduce some notation and capture the above in a formula. We shall work in par amounts of 1 rather than or Variations in the claim amount occur.
For example, deep discount debt may have a claim amount that rises from the issue price to par at maturity, according to some formula or printed schedule. This is not necessarily the case — for example, convertible bonds usually have a low coupon but have a claim amount of par. An alternative claim model is sometimes useful for risk calculations particularly for sovereign debt — see below so we show the formula for this case and make a few further comments.
It is notationally easier to work in continuous time3 but in this case we shall show the formula both in continuous and discrete time. See, for example, Hoskins This is necessary in an implementation of the above model for bond and CDS pricing, and is left as an exercise for the reader. In the typical case — claim amount of par plus accrued — on the default event all debt becomes an immediately due cash amount.
Thus a bond with a one-year outstanding life and a bond with a year outstanding life will both have exactly the same market value after default assuming zero accrued for simplicity — both have the same claim value par , which is due immediately. This is a complicated issue involving the law — usually requiring back taxes to be settled before anything else, and employees to get back pay before banks get repayment of debt, etc. Settlement of claims on debt will be described in the issue document which will usually in the case of corporate and bank borrowers refer to the seniority of debt, and the order in which different seniorities are to be recompensed.
Terminology varies, but debt commonly found in the market and in bank portfolios normally falls into one of three further levels of seniority: loans, senior secured, and subordinated or junior debt.
Once a corporate entity defaults, the administrators of the company seek to realise maximum value from the assets of the company. When the value of these assets is realised then the cash is used in the prescribed order until it is used up.
If this can be met in full, then the remaining cash is applied to senior unsecured bonds and, if these can be repaid in full, it moves on to junior debt. The cash is then used up.
Any deviation from the legal framework and the legally binding issue documents can be challenged in the courts by the creditors. Since these concepts will be used repeatedly, we shall summarise the above in a formula.
We can, of course, introduce more seniority levels as appropriate, but these three are generally all that is required in practice. Recovery as applicable to credit derivative products is a different concept from the above being the one-month post-default bond price and conditions 1. We shall address the question of how we estimate R prior to the default event in the following section.
At present formula 1. Argentinean debt in USD are very similar to those of corporates. Typically the claim amount is also par. Usually there is only one level of seniority for sovereigns.
The major difference between corporate and risky government debt is in the recovery process itself. Compared with ultimate recovery, market recovery data shows relationships of recoveries on different seniorities of debt. Ultimate recovery data will show only at most one seniority with a recovery different from 0 or 1.
However, market prices of defaulted debt are estimates of what ultimate recovery rates will be. Instead of inequality 1. For example post-defaulted assignable loans might be trading at 80, bonds at 50 and junior debt at 20, while ultimate recovery might turn out to be , 65 and 0 respectively. Note that the standard deviation in market recovery rates is less than that of ultimate recovery data — and very substantially less for junior debt.
Secured Sr. Unsecured Sr. Subordinated Subordinated Jr. Subordinated Preferred Stock All Instruments For example utilities tend to have very high recovery rates whereas telecoms tend to be low. Figures 2. A strong relationship between average recovery rate averaged over a large number of defaulted names and seniority of the defaulted debt. Some relationship between industry and recovery rate. Some relationship between rating investment grade versus high yield and recovery rate Figure 2. A dependence on the economic cycle.
The market typically uses the following simple model: Actual values assumed in 2. Given that the impact of the economic cycle on recovery rates is slowly changing Figure 2. We shall see later that these instruments have relatively little sensitivity to the recovery assumption, so the recovery assumption is not important for these instruments. Historical default rates tend to be used by rating agencies and investors in the analysis of portfolios and credit structures.
The historical data gives much more information than just default and recovery rates, which we look at in the following section. We can identify other changes of state — namely transitions from 2 Time-dependent recovery assumptions can easily lead to negative implied default rates. The assumptions made need to be carefully controlled to avoid this and intuitively sensible recovery assumptions may not be consistent with the shape of credit spreads and the requirement to have non-negative implied default rates.
The probabilities associated with these changes of state are given in a transition matrix or TM. Tables 2. AA, A, etc. Aa1, Aa2, etc. For example, from Table 2. The NR column refers to the chance that the name no longer has rated debt at the end of the year — for example, the only rated debt may have matured during the year.
For convenience in subsequent calculations, we have included a row for defaulted names. We assume that once a name has defaulted, it is always in default. In addition, the particular numbers and matrices given are illustrative and not necessarily appropriate for real world applications. But lacking information on how this re-emergence works, we assume that the company stays in default. This is not a bad assumption since — even if the company emerges with no missed payments on debt — in practice the associated recovery value i.
Credit derivatives generally terminate on such events. Rating to: Aaa Aa1 Table 2. Exhibit 13 — Average one-year rating transition rates, — 0. One reasonable assumption would be that, had they been rated, these names would have exactly the same experience as the other names in the sample.
We shall use this assumption in what follows. If the name were risk-free we would apply a discount factor D to the payment to get a current value. However, since the name is risky there is a chance p, say that the name defaults, in which case a recovery amount R is received at the end of the year, say.
We may use various analysis — for example, historical data or detailed company analysis — to make estimates of p and R. We do not necessarily use unadjusted historical data — we are trying to make our best estimates of the real chance of default, and the real recovery in the event of default. These estimates are not necessarily the historical rates — we may explicitly adjust for the state of the economy, known information regarding the company, the state of the property market or other asset markets, etc.
Suppose the same borrower has also made a separate promise to pay 1 in a year, but nothing in the event of default. Suppose that both of these promises are trading actively in the marketplace and that we can observe market prices of P1 for the contract governed by 2. It should be noted that the use of transition matrices does not prejudge which measure we are going to use.
In fact we shall use both natural and implied TMs in following work. Suppose we just have two ratings A and B, plus default. We suppose the 3 by 3 transition matrix is given by Table 2. From 2.
At the end of year 1 the senior debt of the issuer may 6 Actually debt is rated, not the entity. We assume that rating in the TM refers to the senior debt of the company. In this case the senior debt of the entity is rated A. The default probabilities and transition probabilities are given by this matrix and are now taken to refer to the company.
The seniority of the debt we are pricing for that name only impacts the recovery assumptions we are using. A rated Prob 0. B rated Prob 0.
A rated Year 2: B rated Year 2: Figure 2. Repeat the above calculations for a B-rated 2-year bond. This is equivalent to calculating the probabilities of arriving at each node on the tree see Figure 2. Note that as rating changes, the spread jumps from one level to another.
This is an important aspect of the credit market — spread movements in reality can be very large and very quick. Figure Rating or, more generally, perceived quality is certainly a driving factor behind spreads and spread changes.
However we often see spreads changing on a daily basis or a secular change where rating agencies are not changing the rating of individual names. It is also very different from rating in that there are more than seven spread levels in the marketplace. The spread volatility cell E14 will not take into account volatility that arises from changing spreads when rating is constant, so E14 should be an underestimate of the spread volatility in absolute terms.
One way to achieve this is to increase the default probability for a given rating and then multiply the non-default transition probabilities by a factor X so that the sum of the rows remains unity. This can be done in many ways, and one way would be to choose the amount by which you wish to reduce the probability that the rating remains unchanged and increase the probabilities of transition to other ratings collectively by the same amount, distributing the additional amount by increasing these other probabilities by multiplying by a factor.
Note that changes in default and in non-default transition probabilities are not independent, so a manual search for a solution is not easy. Exercise Implement the shift in transition probability described above.
Secondly, the asset swap contract itself is a derivative involving credit risk and, in some versions of the contract, embeds credit risk in a non-trivial way. For further details of asset swap products the reader should consult Das , or Flavell On default of the reference bond the investor receives a cash sum and the contract terminates.
Firstly, the desk pays market price for the reference bond sometimes downloading the bond from the investor. Typically the price differs from par.
Secondly, the desk enters into a contract with an interest rate swap desk where the asset swap desk pays a on day 1 the excess of par over the cost of the reference bond which could be negative — i. The asset swap deal and associated hedges are summarised in Figure 3. Investor Par Figure 3. The contract, as described above, is identical to an investment into par notional of the reference bond plus an interest rate swap, as described above.
It should be noted that, in this form, the investor has full default risk exposure to the reference bond — and maintains economic ownership of the bond. There is no simple short-cut to this calculation. Figure 3. The key point to note is the following: The asset swap spread is one measure of the credit risk on an asset — we shall see alternative measures below z and maturity spread and in Part II CDS premium.
This is a quick and easy calculation. The maturity spread is intuitively a measure of the risk on a bond. The riskier a bond, the less the market will be prepared to pay for it, hence the lower the price, the higher the yield and the higher the spread.
Note that only the bond transaction download of the bond at the market price contributes to the calculation of z-spread or maturity spread. In the case of the Ford bond we note that the z-spread is below the asset swap spread. It should not be a surprise that the asset swap spread is above the z-spread in this case1 since the asset swap is a geared investment in the reference bond. It should be no surprise that the asset swap spread is well below the z-spread as the asset swap is in this case a de-geared investment in the bond.
We describe two and leave their analysis to the reader after Part II has been read.
Asset Swaps and Asset Swap Spread; z-Spread 31 and the desk could sell the bond, unwind the interest rate swap, and make more than par on its hedge. An asset swap with such an embedded option is called a callable asset swap. The reader will see after reading Part II that a callable asset swap can be analysed as a bond download, an interest rate swap, and an option driven by the z-spread on the bond changes in interest rate curve level are assumed to be exactly hedged by the interest rate swap.
On this event unwinding, the embedded interest rate swap contract may lead to a loss because of adverse interest rate moves. This threat can be removed by changing the terms of the asset swap: It is hard to hedge this risk directly — there is an exposure to the reference name s but the size of that exposure depends on interest rate levels. The option can be seen as a swaption but contingent on the default of the reference name.
The integral description is mathematically simple and allows us to concentrate on the underlying features of the model and its implications.
In contrast, a valid implementation must take into account details such as 2 Here we are quoting r and z on a continuous basis.
If we consider several bonds of the same seniority issued by the same entity, how are the z-spreads on these bonds related? Is there a term structure? Is there a coupon structure? How are z-spreads in different currencies related? The model itself does not answer these questions.
The description 3. The model is purely a translation of price into spread, and vice versa. There is no explicit attempt to model seniority differences.
We would certainly expect senior and subordinated bonds for example to trade on different spreads but the model 3.
We might guess that the z-spread curve for various issues has a term structure but no coupon structure. One implication is that, if we are able to observe three bonds of the same maturity but having different coupons, then directly from 3. We would expect the higher priced bond to stand on a higher z-spread. If recovery is, more realistically, anticipated to be above zero then the spread model for pricing bonds cannot be reconciled with a realistic default and recovery model.
A few bonds are special on repo and have yields that do not lie on the surface but are explained by this additional factor. Credit bonds and loans generally do not trade in the same volumes as government bonds — they are less liquid.
Figure 4. Even here the pattern of spreads is not as smooth as we would expect and the deviations cannot be ascribed to repo, seniority or other independently measurable factors. Such differences are usually ascribed to liquidity effects or illiquidity.
The situation worsens if we move away from the major borrowers. Many companies only have one or two bonds in issue: Even if the maturities are similar, an investor would prefer the bond that trades regularly because it is relatively easy to dispose of at a fair price, and there is also a transparent market that allows the valuation of this asset. How much cheaper wider spread should the other bond be? There is no straightforward answer to this question.
Estimating a fair price for the callable given the convertible spread involves modelling in some way equity prices and volatility, interest rates and volatility, and default risk. Suppose only one bond is in issue and rarely trades.
Can we estimate the bullet spread, and how? Looking at marks provided by credit traders may not give us the answer. Front book credit bond traders make markets in bonds and will attempt to make a market in illiquid issues and names.
If possible they will attempt to match a seller with a downloader on such issues. A large bank may have a portfolio of debt issued by 10 or more entities. This might 34 Credit Derivatives 80 60 40 20 0 0 2 4 6 Life 8 10 12 Figure 4. Of the traded issues, perhaps trade regularly in the bond market, perhaps another trade irregularly in the bond and secondary loan market, and the other trade rarely.
This pattern is illustrated in Figure 4. Marking such a book to market would present considerable problems. But suppose the bank wishes to clear the economic risk associated with a large part of its portfolio off its books by issuing a derivative related to the reference entities. Investors in the derivative who are synthetically downloading the underlying risks will need to know the current spread associated with each of the names.
Other maturities present a problem. The bulk of the portfolio does not trade. Some products combine both of these methods. The spreads estimated may be bond spread or default swap premium rates — the principles and methods used are nearly identical. We describe products for default swap data below. The contributing individuals are effectively acting as market-makers and are trying to come up with a spread bid and offer on which they would actually be prepared to deal.
As such, a good market-maker will look at various sources of information — known deals, how the market has moved, how this company compares with other similar companies, ratings and changes, current company news, etc. Given certain points on the maturity spectrum, produce spreads for all other maturities.
Given a historic spread for the name, estimate where it is likely to be trading now, given data on the general movement of spreads over the intervening period. Such approaches are used by several data providers. The starting point is to collect data for traded names and calculate average spread and the standard deviation — see Table 4.
Also there will be considerable Table 4. Such a process of estimation is only appropriate for derivatives based on large portfolios of names e. Counterparties which do not meet these criteria will generally not be accepted as counterparties without other conditions being met — for example, full collateral being posted or being accepted for certain types of trades or maturities only. Even once a counterparty has been deemed an acceptable risk, further measures described below are taken to control the risk on individual deals.
Thus a mature deal — for example, an interest rate swap or default swap which had an initial value of zero — may have moved to being USD 2m positive value for party A and the opposite for party B.
Under the collateralisation agreement party B will have paid party A USD 2m collateral on which party B receives interest. If party B defaults, and the interest rate swap contract vanishes, then party A has 2m capital which it can use to replace the deal with another counterparty.
To be effective, collateralisation also requires a netting agreement — in the event of default by party B the net value of all the trades with party A some of which have a positive mark to market, others have a negative mark-to-market only is a claim of one on the other. Note that collateralisation does not eliminate the counterparty risk for two reasons: This risk may be positive or negative.
Also in practice the failure of a counterparty may take some time to establish. A counterparty may fail to post collateral one day because of an administrative error. If the counterparty is failing it will probably try to maintain trading relationships as long as possible — and make excuses why collateral has not been posted. In practice it takes between about 3 and 10 days to determine that a 40 Credit Derivatives counterparty has failed and to close a deal.
If the underlying contract is a credit derivative there may be a correlation between the counterparty and the reference name s in the credit derivative which is commonly the case for portfolio credit derivatives. The default of the counterparty will be associated with a widening of the spread on the reference names and a substantial change in the value of the deal. Default of the counterparty may be related to a jump in the spread on the reference name s , hence a step move in the value of the underlying deal when the counterparty defaults.
It is common to require that there is no correlation between the counterparty and the reference deal on credit derivatives, in addition to the other counterparty requirements described above. Assessing whether correlation exists between a reference entity and the counterparty is usually a subjective judgement and is typically based on some or all of a region and industry overlaps b direct and indirect business relationships c equity correlation.
The more certain deals use up the allocated capital, the fewer such deals are done in favour of less counterparty capital-intensive deals.
Detailed understanding of VaR approaches are not needed for this section — Jorion or Grayling provide much more detail. We can calculate a forward value of that asset but are interested in how far the value can fall below the expected value. More generally we are interested in the distribution of market values, and the low-value tail of this distribution. We shall refer to this concept as VaR — we shall suppose that we measure this as the deviation of value below the expected value corresponding to a certain percentile of the distribution of value.
Our interest is in credit risky assets — the term creditVaR is sometimes used to refer to the risk to the value arising from changing credit spreads or defaults.
A trade may have counterparty risk, for example, and interest rate swap or a credit default swap has exposure to a counterparty.
We can calculate the expected exposure to the counterparty at a forward date. But, again, our interest may be in how great this exposure to the counterparty could be. In this case we are interested in the high-value tail of the distribution of the asset value at the forward date. We refer to this concept as counterpartyVaR. For both measures we need to be able to produce a distribution of forward values of the underlying asset.
Typically we are not interested in the VaR or counterpartyVaR numbers on a particular deal. We are much more interested in the VaR on the entire portfolio of assets, or the counterpartyVaR to a particular counterparty for all trades related to that counterparty.
For example, let us suppose that we own a 5-year FMC bond priced at par. Then this will have a certain VaR — say X. If — when thinking about the two deals together — the bond falls in value, the insurance will rise in value since the combination of the two will always be worth par see Part II. So in this case the VaR of the pair of deals is zero. We cannot add the VaR numbers for individual deals in order to get this: We can calculate the expected forward price of a bond of this name using the transition matrix as described in Sections 2.
We shall do something extra here because we want to be able to price the bond in the presence of other reference entities whose transitions are correlated. For example, it might contain autos, telecoms, consumer good names, etc. Similarly, consumer goods and autos will show some correlation because of general economic factors, while utilities and brewing will be much less correlated. We therefore wish to build in the idea of correlated transitions.
First we shall create correlated transitions; then, once we have the forward rating for each entity, we use the forward rating and price the bond using the previous method. If we choose the latter, then the TMs are calibrated to replicate the average spread by rating the bonds in the actual portfolio held, rather than market average spreads. Once we have chosen the transition matrix we wish to use, we can simulate the forward prices of the bond as follows. Pick a random number x from a normal distribution.
Compare this probability with the probability from the transition matrix that the A bond has migrated to 1.
Figure 6. We can use this method to simulate the price of the bond at the end of the year knowing the simulated rating and the forward life we can calculate the spread using the TM approach described in Chapter 2 and hence the price.
Of course if we are using the natural measure then the prices themselves are unrealistic — but typically we are looking at the distribution of price changes and we apply those changes to the actual market price. Two random numbers from the uniform distribution are taken a standard Excel function , one is converted to an exponential distribution, and then a simple transformation produces two independent numbers for the normal distribution.
Credit Portfolios and Portfolio Risk 43 0.
Normal distribution and areas corresponding to rating transition probabilities If we use the risk-neutral measure we not only incorporate accurate bond pricing but also a volatility measure for bond spreads based on for example actual spread volatility rather than just volatility arising from rating transitions.
We can now use this simulated forward bond price distribution for VaR calculation. Similar techniques can be used for credit derivatives — and we can also use this method to calculate the distribution of counterparty exposure on a credit default swap at a forward date. If we need to simulate over an n-year period then we have two choices: The latter approach is usually adopted because of speed.
This is found to be trivial: How do we produce correlated normal random numbers? There are various ways: The Cholesky matrix is a triangular matrix with the following property: So, if we generate two independent uniform random numbers x1 and x2 then the two numbers x1 and 0. The extension to many names is mathematically trivial but is best handled in code rather than a spreadsheet.
The process described above — generating correlated Normal random numbers using a correlation matrix and independent Normal random numbers — is referred to as the Normal or Gaussian Copula and is discusses in detail in Part III. Imaging that the random variables are three coins which may come up heads or tails. If A comes up heads then B is very likely to come up heads, and C is also very likely to come up heads. Likewise, if A comes up tails, B and C are both very likely to come up tails.
So in most experiments all three coins come up with the same face showing. The correlation between B and C has to be high — it turns out that 0. The implication is that the correlation matrix cannot be set up arbitrarily, and cannot be stressed arbitrarily. Once we have the forward rating we calculate the forward price using the method of section 6. The missing step so far is: Names that have the same rating at the start of the period will therefore have the same rating at the end.
Observing the transitions of names over many years is not likely to lead to useful results since, in particular, default transitions are rare and a very large number of years would be required. There are several approaches to correlation: We could consider each name by name pair and calculate a correlation based on some data history. One approach commonly adopted is to use the equity price correlation matrix.
We introduced the idea in section 6. Here we are using industry as a tag to group companies together rather than considering a name-by-name correlation. Another approach is to choose correlations arbitrarily usually driven by tags — e.
In the context of VaR calculations it is common to use correlations derived from equity data. The process of derivation of those correlations is typically not pairwise but based on a factor approach.
We shall revisit these topics in a different context in Part III. Likewise we shall regard credit bonds and loans as credit derivatives. We shall present models which cover these instruments as well as the more obvious credit derivatives. The traditional credit market covers many more instruments than simply bullet debt. Callable or puttable debt follows the same pattern as for the non-credit market. Such bonds require a model of interest rates and credit spread and default risk.
In addition, convertible debt gives an option to exchange the bonds for a certain number of shares. These add a third dimension to the model which also requires equity prices to be modelled. Commercial banks have been offering a variety of derivatives of varying complexity for nearly as long as they have been granting loans.
It is economically the same as a spread option on a bond. The commercial banking forms are often not priced in the same way, are generally not traded but held to maturity or expiry, and do not require a mark-to-market value — unlike the traded equivalents. It is similar to insurance on the debt of the company, with the main differences that it is not an insurance policy and there is usually a range of deliverable debt. The CDS forms the core of the credit derivative business in terms of numbers of deals done.
Spread options arise in a variety of forms. A typical example is the right to sell a bond at a certain spread over a reference rate at a certain time in the future. Instead of a bond the underlying instrument may be a default swap. Callability or puttability in bonds is usually more complicated, driven by both interest rate levels and spreads.
Effectively the underlying traded asset is exchanged to an off-balance sheet asset, which is economically equivalent. A total return swap is very similar to a repo trade.
The note may terminate early, and repay less than par, on a trigger credit event of a reference entity or entities.
The simplest example is where a single name default swap is repackaged into a CLN, but the structure may be much more complicated than this, for example, being related to the risk on a mezzanine tranche of a CDO.
A single contract describes a collection of default swaps, but otherwise the two deals are identical. An average basket is actually a single tranche CDO. Examples are the iTraxx and sub-index portfolio CDSs. Nth to default baskets are quite different products. A payment is made in return for defaulted debt on a trigger event for any one of a pre-agreed list of reference names — usually between 3 and The contract then terminates.
The nth to default basket is becoming a vanilla product among portfolio credit derivatives. The product often occurs when a commercial bank seeks protection on its portfolio of loans. CDOs occur in many forms. The iTraxx Europe index relates to a standard portfolio of , and there are over 10 smaller indices from 10 to 30 names representing industry groups, sub and senior CDSs, high-yield and other sets of names.
Market-makers trade standard portfolio structures based on this — generally a single tranche CDS subject to premium which is close to the average CDS premium when the portfolio started. In addition they trade a standard indexing structure on the name reference portfolio. Portfolio spread options have also recently been introduced. The presence of actively traded tranches of standard CDOs has opened up the possibility of trading spread and spread volatility on a standard benchmark.
The portfolio spread option gives the downloader the right to download or sell the underlying CDO tranche at a predetermined tranche premium. Introduction to Credit Derivatives 49 7. Figure 7.
It should be borne in mind when looking at market data on credit derivatives that the underlying product is a structured product that is not traded through an exchange.
There is no independent source of volume of transactions or size and type of deals, so data, particularly from the earlier years when the market was less transparent, has to be viewed with caution. Table 7. These products form the core of the CD business, are the vanilla trading products, and form the core of the credit instruments that are used to hedge more complicated credit structures.
Credit-linked notes are also largely embedded single name risks. Note that embedded credit derivatives, in particular spread options embedded in other products, are not captured by the survey. The reference entity in the transaction may be a corporate including banking entity or a sovereign.
In the case of a sovereign reference entity the default risk typically refers to obligations issued in a currency other than that of the sovereign.
Some traders and investors use the same terms to refer to downloading or selling risk the opposite position — it is obviously important to be clear on the terminology you and your counterparties are using. The quality of the counterparty to the deal has risk and pricing implications but note the contents of section 8. We can see, before looking at details of the contract, that the deal is a portfolio deal: We shall examine the pricing impactions of this later, and in most of this part we shall assume that both the writer and the downloader are risk free.
The contract typically pays par in return for nominal of debt if the reference entity suffers a credit event before the maturity of the deal. The downloader typically pays a premium quarterly in arrears with a proportion up to the default date of the reference name in the event that default occurs before the maturity of the trade. Section 8. DB Figure 8. FMC Protection downloader: Post-default Figure 8.
Note that delivered debt may be YEN, EUR or other currency debt, but the notional amount of debt is chosen to be the same as USD 10m, taking the current exchange rate on the day that notice of the bond to be delivered is given. This is called the delivery option, and was originally introduced to reduce the risk of a squeeze developing on a deliverable issue.
A credit event would cause that particular deliverable to rise to par. If the bond could be bought below par, the holders of protection with no debt to deliver could download the bonds to deliver and obtain the capital gain on the CDS; holders of debt and CDSs would have no incentive to sell debt below par, and holders of bonds only would be bid up until the price reached par.
CDS premia are often referred to as spreads, which is misleading since they are not a spread to anything the terminology arose because a CDS premium — expressed in basis points per nominal — is of similar magnitude to a bond spread. Wrapped bonds have many of the features of a vanilla CDS contract. Average baskets refer to a portfolio of names and are documented in a very similar way to a single name CDS. Each name may have a notional amount and premium rate associated with it. On a credit event on one name, the notional amount is paid in return for defaulted debt, the premium associated with that name ceases, and the basket continues with the remaining names.
In this case an average basket is merely a portfolio of single name CDS contracts, the only difference being that a single piece of documentation is signed rather than many. Average baskets formed the predecessor to standardised portfolio credit derivatives — such as the products based on the iTraxx indices.
The iTraxx-based deals are mostly of this form. Bank guarantees are traditional contracts granted by commercial banks to certain customers. Economically they are similar in form to CDS contracts.
This documentation represents a format that can be varied to a greater or lesser extent 58 Credit Derivatives and has changed over time — so many older contracts on the books of traders and investors have slightly different documentation.
At one time the term default option was used for deals where the CDS premium was paid up-front. Insurance Contracts and Documentation A CDS is economically very like an insurance contract on debt issued by a company. What are the differences between a CDS and an insurance policy? The key differences are as follows: An insurance contract requires the owner of the insurance to own the insured risk at the time a claim is made.
In addition, insurance companies — the writers of insurance policies — are regulated by an insurance regulator; banks are regulated by a central bank; and other bodies may not have a regulator other than generally through accounting and legal requirements.
Insurance documentation may look very different from ISDA documentation. Materially different conditions may apply — for example, replacement clauses for reference entities may exist in certain circumstances e.
They are then not insurance contracts but simply a CDS where the counterparty is an insurance company. An insurance contract written under ISDA vs a banking contract.
Minimal — the counterparty is strictly regulated. A CDS written by a hedge fund vs a banking contract. Unregulated counterparty, probably unrated. A CDS written by a corporate vs a banking contract. As above, though large corporates typically have rated debt.
How would you control the risk in the above?
For example, a CDS traded on 5 February may have had an effective date of 12 February and a maturity date of 12 February This means that a new 5-year deal will have anything from 5 to 5. Single up-front or at maturity premia also occur. In these cases a proportion may or may not be refunded paid on early default.
This is usually related to a commercial bank trade where the bank wants to get protection for a particular asset on its books. At one time regulatory treatment was unclear, and specifying the deliverable was an attempt to prove to an uncertain regulator that the protection did in fact cover the risky asset, and regulatory relief could then be obtained. Cash or Physical Cash settlement is an alternative possibility to physical settlement.
On a credit event the writer pays par less the market value of defaulted debt of the reference name less any accrued premium and subject to a minimum of zero. This typically involves a dealer poll to get independent valuations of defaulted debt of the appropriate seniority and taking an average price.