4 Swaps and Credit Derivatives
4 Swaps and Credit Derivatives
2.niapS
Table of contents
1. Swaps* ........................................................................................................................... 1
1.1 Definition and characteristics* ...................................................................................................... 2
1.1.1 Interest rate swaps * ...................................................................................................................... 4
1.1.2 Swap quotes*................................................................................................................................. 6
1.1.3 Closing a swap position* ............................................................................................................... 7
1.1.4 Other interest rate swaps * ............................................................................................................ 7
1.1.5 Currency swaps * .......................................................................................................................... 8
1.1.6 Other currency swaps * ............................................................................................................... 11
1.2 Strategies using swaps * ............................................................................................................... 12
1.2.1 An illustration: fixed for floating interest rate swaps * ............................................................... 12
1.2.2 An illustration: fixed-for-fixed currency swap * ......................................................................... 15
1.2.3 Another illustration: fixed-for-floating currency swap * ............................................................. 16
1.2.4 Strategies and risk-management with swaps * ............................................................................ 17
1.2.4.1 Creating synthetic fixed or floating rate liabilities* ................................................................................ 17
1.2.4.2 Hedging foreign currency exposures* .................................................................................................... 18
1.2.4.3 Asset based swaps*................................................................................................................................. 19
1.2.5 Swaptions and swap futures* ...................................................................................................... 20
1.3 Pricing and valuing swaps* .......................................................................................................... 21
1.3.1 Pricing of interest rate swaps* ..................................................................................................... 21
1.3.1.1 Pricing a swap as a portfolio of bonds* .................................................................................................. 21
1.3.1.2 Pricing from a zero-coupon yield curve (forward contracts)*................................................................. 23
1.3.1.3 Pricing of currency swaps* ..................................................................................................................... 25
1.3.1.4 Pricing a currency swap as a portfolio of bonds* ................................................................................... 25
1.3.2 Pricing a currency swap using forward contracts * ..................................................................... 26
1.4 Other types of swaps* ................................................................................................................... 27
1.4.1 Equity swaps*.............................................................................................................................. 27
1.4.2 Commodity swaps* ..................................................................................................................... 27
1.4.3 Volatility and variance swaps* .................................................................................................... 28
1.4.4 Credit default swaps* .................................................................................................................. 29
1.4.5 An introduction to financial engineering* ................................................................................... 29
2. Credit derivatives: market, instruments and general characteristics* ................. 31
2.1 Market of credit derivatives* ....................................................................................................... 31
2.2 Credit default swaps (CDS)* ........................................................................................................ 33
2.2.1 Definition*................................................................................................................................... 33
2.2.2 CDS and bond yields* ................................................................................................................. 34
2.2.3 Credit events* .............................................................................................................................. 36
2.2.4 Settlement* .................................................................................................................................. 37
2.2.5 Credit default swap: Bloomberg example* ................................................................................. 38
2.2.6 Index CDS products* .................................................................................................................. 40
2.3 Credit linked notes (CLN)* .......................................................................................................... 43
2.4 Other credit default swap products* ........................................................................................... 47
2.5 The role of credit derivatives* ..................................................................................................... 48
2.5.1 Isolating credit risk* .................................................................................................................... 48
2.5.2 Efficient mechanism to short a credit* ........................................................................................ 48
2.5.3 Market for pure credit risks* ....................................................................................................... 51
2.5.4 Liquidity provision in times of turbulence* ................................................................................ 51
2.5.5 Tailoring credit investments and hedges* ................................................................................... 51
2.5.6 Confidential transactions* ........................................................................................................... 52
2.6 Market participants* .................................................................................................................... 53
3.niapS
2.6.1 Bank and loan portfolio managers* ............................................................................................. 53
2.6.2 Market makers* ........................................................................................................................... 53
2.6.3 Hedge Funds* .............................................................................................................................. 54
2.6.4 Asset Managers*.......................................................................................................................... 54
2.6.5 Insurance companies* .................................................................................................................. 54
2.6.6 Corporations* .............................................................................................................................. 54
2.7 Institutional framework* .............................................................................................................. 55
2.7.1 Marking to market* ..................................................................................................................... 55
2.7.2 Standardized documentation* ...................................................................................................... 55
2.7.3 Counterparty consideration* ........................................................................................................ 56
2.8 Spread volatility of credit default swaps* ................................................................................... 57
2.9 Credit derivatives: valuation of credit default swaps* ............................................................... 58
2.9.1 Absence of Arbitrage: Creating synthetic CDS* ......................................................................... 58
2.9.2 Valuation of credit default swaps by a non-arbitrage approach* ................................................. 59
2.9.3 Estimating default probabilities* ................................................................................................. 62
* final level
4.niapS
Derivatives
1. Swaps*
Of all the over the counter financial innovations introduced in the 1980s, none can rival the
swap market. Its growth over the past 20 years has been phenomenal. Today, capital market
swaps are one of the most important classes of what are known as derivative instruments.
Swaps compete with, and at the same time complement other risk management tools such as
futures and options.
5.niapS
cash flows 2
Counterparty 1 Counterparty 2
cash flows 1
At the early days of the swap market, transactions were arranged by banks and investment
banks, which were compensated with a fee. This meant the counterparts to the swap signed
the swap contract with each. Today, it would be very difficult to arrange a swap directly
between two end-users. The problem can easily be solved by employing a financial
intermediary. In exchange for a commission, swap brokers find counterparties with matched
needs and then negotiates with each on behalf of the other. However, they assume no risk as
they do not take a position in the final swap. By contrast, swap dealers or market makers
are willing to act as the counterparty in a swap. Assuming that they do not want to bear the
associated risks, they will need to find a matching counterparty or hedge their book. The swap
dealer profits from the pay-receive (or bid-ask) spread it quotes on the swap flows.
Using a swap dealer is advantageous for both parties for two reasons:
• The use of an intermediary reduces search time in establishing a swap agreement. The
swap dealer is ready to enter into a swap at any time, whereas both counterparties might
take several days to discover each other, even with a broker's help;
• An intermediary can reduce the costs of credit evaluation. Each of the participants in an
interest rate swap may become bankrupt and unable to fulfil their side of the contract. As
the two counterparties are not familiar with each other, each would need to undertake
costly credit analysis on the other before agreeing to deal directly.
6.niapS
The cash flows associated with a typical swap are illustrated in the following figures, which
depict the initial exchange of notional (which is optional in the sense that it is not required in
all swaps), the periodic payments between the counterparties, and the re-exchange of notional
(which is also optional). For the sake of simplicity, we have omitted the bid-ask spread on the
periodic payments.
notional 2 notional 2
Counterparty 1 SWAP Counterparty 2
DEALER
notional 1 notional 1
Flow 2 Flow 2
Counterparty 1 SWAP Counterparty 2
DEALER
Flow 1 Flow 1
notional 1 notional 1
SWAP
Counterparty 1 Counterparty 2
DEALER
notional 2 notional 2
In contrast to most other financial markets in the United States, the swap market is subject to
remarkably little regulation, and does not have a central exchange, not even a central clearing
mechanism. The terms of a swap agreement are determined by the parties to the contract, even
though the majority of them use document forms suggested by the International Swap Dealer
Association (ISDA) or the British Banker’s Association 1. Note that the exact terms or even
the existence of a swap does not need to be disclosed and reported at the time the agreement is
executed. So far, this lack of regulation has not resulted in major problems.
Note that even though the swap market is not subject to regulation, the individual participants
are. In particular, federal banking regulators in the United States have included swaps in the
risk-based capital standards. Swaps are off-balance sheet transactions, that is, they do not
appear on the asset side or on the liabilities side of a balance sheet, but they have to be
considered when assessing the total risk exposure of a financial institution.
1 The ISDA publishes a book of definitions and terms to help standardize swap contracts.
7.niapS
There are three major types of participants on the swap markets: end-users, intermediaries,
and bookrunners:
• the end-user engages in a swap for economic or financial reasons. He wants to obtain low-
cost financing (liability) or obtain high-yield investments (asset), hedge his currency and/or
interest rate risk, or maybe speculate;
• the intermediary engages in a swap to earn fees. It provides credit-enhancement to swaps,
matches counterparties, and earns a fee-based income;
• the bookrunner engages in a swap to earn trading profits. He views swaps as a tradable
security, makes the prices, promotes liquidity, and may take trading positions.
The most common type of interest rate swap is a fixed-for-floating rate swap, also called
plain vanilla swap, or generic swap, in which the payments to be made by one party are
calculated using a floating rate of interest, while the payments to be made by the other party
are determined on the basis of a fixed rate of interest. Payments are usually made in arrears on
an annual or semi-annual basis. The effective date is the date interest begins to accrue, and
the payment date is the date interest payments are made.
This can be illustrated by the following figure. In this example, company ABC agrees to pay
company XYZ a fixed rate of interest. In return, XYZ agrees to pay ABC a floating rate of
interest 2.
The floating interest rate is periodically reset, that is, it is pegged to some specific spot
market rate (called the reference rate), which is observed on specific reset dates. The most
commonly used floating rate is LIBOR, which is the interest rate quoted for Interbank
lending of Eurocurrency deposits (the acronym of London Interbank Offered Rate). There is a
LIBOR rate for USD deposits, for EUR deposits, CHF deposits and so on.
8.niapS
The fixed-rate payer is referred to as the payer. He can also be called the floating-rate
receiver and is often referred to as having “bought” the swap or having a long position. The
floating-rate payer is referred to as the receiver. He can also be called the fixed-rate receiver
and is often referred to as having “sold” the swap and being short.
The two payments are called the legs or sides of the swap. The fixed rate is called the swap
coupon. The payments are calculated on the basis of a hypothetical quantity of money, called
the notional amount, or notional principal, which is usually not exchanged in an interest
rate swap.
Example:
On 28.12.Y–1, company ABC enters in a swap transaction with company XYZ. Both agree to
swap from the 01.01.Y, cash flows on a notional of CHF 100 Mio. until the 31.12.Y+4 (5 years).
Company XYZ (the payer) will pay to company ABC once a year a fix rate of 4% on a notional
amount of CHF 100 Mio. Company ABC (the receiver) will pay every six months a floating
interest payment of half the annualised six-month LIBOR of the same notional amount of CHF
100 Mio. (LIBOR setting). Effective dates are 01/01 and 01/07.
From ABC’s point of view, at the beginning of the swap, the cash flows are the following 3:
Note that LIBOR is observed at the beginning of the period. Due to this, the floating rate for the
initial calculation period is known on the trade date and is usually included in the swap
confirmation. As the future LIBOR rates are unknown, the future floating-rate cash flows to be
paid are also unknown. If Libor turns out to be greater than 4% p.a., then the fixed payer will
receive the interest differential. If LIBOR < 4%, then the floating payer will receive the interest
differential.
Note that in practice, it is very important to understand the day-count conventions that apply
to a swap. The fixed and floating rates are often not directly comparable. For instance, the
floating leg generally follows money market convention in the relevant currency (Act/360 or
Act/365), while the convention for the fixed leg generally follows the bond convention in that
currency/region (e.g. 30/360 for U.S. bonds).
3 Here, for simplicity, we have not considered the effective number of days and we have assumed that LIBOR
setting was made at the payment date.
9.niapS
Swap prices are frequently quoted as a spread over government issues, therefore serving as a
rough indicator of credit risk of the banking sector. An example of a schedule for interest rate
swaps of various maturities is given below.
This schedule assumes semi-annual rates and bullet transactions (i.e. non-amortising) for top
credit firms. All rates are quoted against six-month LIBOR flat. For each maturity/tenor, we
need to add the quoted spread to the yield on the “current” Treasury note of the same
maturity. For example, if the dealer pays fixed on a 5 year swap, the fixed rate of the swap
would be equal to the yield on the 5 year note plus 52 basis. If the dealer pays floating and
receives fixed, the fixed rate of the swap would be equal to the yield on the 5 year note plus
60 basis points.We can obtain mid-rate by simply taking the average of the paid and received
rates. For example, the five-year interest rate swap is 6.48 percent, which is calculated as:
A popular benchmark in the market is called the swap curve. It plots the fixed-rate leg of a
plain vanilla swap [i.e fixed against six-month LIBOR] for various maturities. Note that its
construction is still not a uniform practice, due to substitutable inputs, overlapping instrument
maturity dates, inconsistencies between different inputs, use of various interpolation
techniques, etc. Nevertheless, the swap curve remains a pivotal element in pricing fixed-
income products, measuring the relative value of debt classes, and measuring interest rate
expectations.
4 Note that LIBOR is an unsecured rate – it is the short term rate at which major international banks can
borrow unsecured funds from each other, that is, without posting collateral.
10.niapS
Example:
A fixed rate swap payer sees interest rates decreasing. As a result, he wants to change his position.
What are the options available to him?
For instance, the parties may decide to exchange a floating rate against another floating rate,
say, for example, LIBOR against prime rate. This allows financial institutions to hedge an
exposure arising from assets and liabilities that are subject to different floating rates. Swaps in
which both legs are floating rates are called basis swaps. Yield curve swaps are basis swaps
in which the counterparties agree to exchange payments based on the difference between
interest rates at two points on a given yield curve.
Some swaps do not use a fixed principal amount: amortising swaps are swaps on which the
notional principal is reduced at some point(s) in time prior to the termination of the swap.
Accreting swaps are swaps on which the notional principal is increased at some point(s) in
time prior to the termination of the swap. Roller coaster swaps are swaps that include both
accretion and amortisation of the notional principal. Mortgage-indexed and collateralised
Mortgage Obligation (CMO) swaps provide for the amortisation of the notional in a manner
consistent with the amortisation of a mortgage or a CMO pool.
Zero-coupon swaps are fixed-for-floating swaps in which the fixed rate is a zero-coupon (no
payments are made for the fixed rate until maturity, at which time a large payment is made).
Forward swaps or deferred swaps are swaps in which the swap coupon is fixed on the
transaction date, but the swap will not start accruing interest until a later date. By opposition,
delayed-rate setting swaps are swaps that commence immediately, but on which the swap
coupon is not set until a later date.
11.niapS
Reversible swaps are swaps on which the payer and the receiver reverse roles one or more
times during the swap life, that is, the fixed rate payer pays a floating rate, and the floating
rate payer pays a fixed rate.
Differential swaps are swaps on which interest rates are set in one single currency, but where
the floating rate is indexed to a foreign currency interest rate.
Callable swaps (similarly putable swaps) are swaps on which the payer (the receiver) has
the option to terminate the swap early. Extendible swaps are swaps on which one party has
the option to extend the maturity of the swap.
A currency swap is very similar to an interest rate swap, except from the fact that:
• There are two currencies involved; thus, the notional amounts and the cash flows that are
exchanged are labelled in two different currencies.
• The parties exchange interest payment streams of different nature: fixed against fixed,
floating against floating, or fixed against floating.
There are generally an initial and a final exchange of the principal amounts 6. Typically,
principal amounts are exchanged at the start date of the swap and at the maturity date of the
swap. Hence, the basic currency swap involves three distinct sets of cash flows:
• the initial exchange of principals, labelled in different currencies, which sets the
exchange rate between the two currencies
• the interest payments made by each counterparty to the other
• the final exchange, or re-exchange, of principals, on the basis of the spot exchange rate
prevailing at the inception of the swap.
The cash flow structure of a “plain vanilla” currency swap is illustrated by the following
figures.
5 A cross-currency swap has not to be confused with a traditional foreign exchange swap, which is simply a
spot currency transaction that will be reversed at a predetermined date with an offsetting forward transaction;
the two are arranged as a single transaction.
6 Hence, it is not really correct to call the principals on a currency swap “notional” since the principals may be
actually exchanged while the accepted meaning of the term notional implies “hypothetical”.
12.niapS
currency 1 currency 1
principal principal
Counterparty SWAP Counterparty
1 DEALER 2
currency 2 currency 2
principal principal
floating or floating or
fixed rate in fixed rate in
currency 2 currency 2
Counterparty SWAP Counterparty
1 DEALER 2
floating or floating or
fixed rate in fixed rate in
currency 1 currency 1
currency 2 currency 2
principal principal
Counterparty SWAP Counterparty
1 DEALER 2
currency 1 currency 1
principal principal
13.niapS
Example:
On 28.12.Y–1, company ABC enters into a swap transaction with company XYZ. Both agree to
swap from the 01.01.Y cash flows on a notional of 100 Mio. CHF at a 4% fixed rate until the
31.12.Y+4 (5 years) and cash flows on 80 Mio. USD at the USD 6-month LIBOR.
Company XYZ will pay to ABC once a year a fixed rate of 4% on a notional amount of
100 Mio. CHF. Company ABC (the receiver) will pay every six months a floating interest
payment of half the annualised six-month LIBOR of the notional amount of 80 Mio. USD (LIBOR
setting).
In this example, the spot USD/CHF exchange rate would be 1.25. On December 31, N + 4,
company XYZ pays back ABC 100 Mio. CHF and receives from ABC 80 Mio. USD regardless of
the spot exchange rate at that time.
From ABC’s point of view, at the beginning of the swap, the cash flows are the following 7:
As the future LIBOR rates are unknown, the future USD floating-rate cash flows are also
unknown, except for the first payment.
7 Here, for simplicity, we have not considered the effective number of days, and we have assumed that LIBOR
setting was made at the payment date.
14.niapS
Fixed-for-fixed rate currency swaps are swaps in which both counterparties pay a fixed rate
of interest. Note that they can be created via a single swap agreement, or through the
combination of two fixed-for-floating swaps using the same floating rate. Such swaps are also
called circus swaps.
Floating-for-floating currency swaps (often called basis swaps in the case of LIBOR-
LIBOR swaps) are swaps in which both counterparties pay a floating rate of interest. Here
again, note that they can be created via a single swap agreement, or through the combination
of two fixed-for-floating swaps using the same fixed rate.
Amortising currency swaps are swaps in which the principals are re-exchanged in steps.
Accreting currency swaps are swaps in which the principal is scheduled to increase over the
life of the swap.
15.niapS
Let us consider the following example: two companies, A and B, want to borrow from the
market. The offered market conditions are the following:
Not surprisingly, company A has a lower credit risk and therefore can obtain better conditions
than B for both rates. Company A has an absolute cost advantage versus company B in raising
funds in either the fixed or floating debt markets, but company B has a comparative
advantage 8 compared to A borrowing at a floating rate. The difference in quality spread, or
the quality spread differential, is 75 basis points in this example.
This quality spread differential may be exploitable if and only if company A wants to borrow
at a floating rate, while company B seeks a fixed rate (where it has a comparative
disadvantage). In such a case, a swap could be made as follows:
Step 1: Initial borrowing of principals: Each company issues debt on the market in which it
has the greater comparative advantage. Company A will borrow at a fixed rate of 3%, and
company B will borrow at a floating rate of LIBOR + 0.50%. To simplify, we will assume
that both loans have a time to maturity of five years and a principal amount of CHF 100 Mio.
8 Or a “lower disadvantage”!
16.niapS
Counterparty A Counterparty B
Step 2: Enter into the swap transaction: Then, A and B enter in a plain vanilla interest rate
swap agreement in which they agree (for example) on the following conditions: 5 years, CHF
100 Mio notional, and the two swapped rates are 3% versus LIBOR flat.
Step 3: One year later: Let us consider what happens for the debt and swap service
payments. We will assume that the one-year LIBOR was equal to 3.50%.
For the swap service, A has to pay to B an amount of CHF 3.5 Mio corresponding to
Libor=3.50% on the notional amount of CHF 100 Mio., while B must pay to A an amount of
CHF 3 Mio corresponding to a fixed rate of 3% on the notional amount of CHF 100 Mio. In
practice, in an interest rate swap, only the net cash flow would be exchanged, i.e. A pays to B
CHF 500’000.
For the debt service, A has to pay an amount of CHF 3 Mio corresponding to a loan of CHF
100 Mio at 3%, while B has to pay an amount of CHF 4 Mio corresponding to a loan of CHF
100 Mio at LIBOR + 0.50% = 4%).
LIBOR %
Counterparty A Counterparty B
3%
17.niapS
• company A receives a 3% fixed rate from B, and uses it to pay the interest of its 3% fixed
rate borrowing on the market. A pays a LIBOR floating rate to B. The result is that A
borrows at a floating rate of
LIBOR + 3% – 3% =
LIBOR
while it would have paid LIBOR + 0.25% if borrowing directly at a floating rate. This
yields an interest rate gain of 0.25%.
• company B pays a 3% fixed rate to A, and receives a LIBOR floating rate from A. Then, B
pays a (LIBOR + 0.50%) rate on the amount borrowed on the market. The result is that B
borrows at a fixed rate of
Through the swap, both companies have lowered their effective borrowing interest rate.
We can note that the total interest rate gain (0.50% + 0.25%) is equal to the interest rate
spread difference of the two companies (0.75%). Note that in this example, the split is
arbitrary. The two parties could split the gains differently – this is essentially determined as a
result of the negotiation of the swap terms. However, the total gains to the swapping parties
will always equal the quality spread differential: 75 basis points in this example.
Step 4: At maturity, both initial loans terminate, and each counterparty has to pay back the
principal amount of CHF 100 Mio. 9.
Counterparty A Counterparty B
What are the underlying reasons for the existence of quality-spread differentials? Some have
argued that quality spread differentials are due to market inefficiencies. But if that was really
the case, the annual volume of new swaps should be declining as arbitrage becomes more
effective. Various other theories explain that quality spread differentials could arise from
differences in expected bankruptcy costs, agency costs, various interest risk exposures, etc.
9 In addition, there is also a swap of interests, and an interest payment to the lenders, but the principle has been
exposed in the previous figure.
18.niapS
Suppose that a British firm needs a fixed-rate USD funding, while an American firm needs
fixed-rate GBP funding, both for a period of five years. The market conditions for both firms
are the following:
It is clear that the British firm has an absolute borrowing advantage in both GBP and USD.
But the American firm has a comparative advantage in the USD market.
Since the American firm needs fixed-rate GBP funding (where it has a comparative
disadvantage), a swap could be arranged as follows:
Step 2: The two firms enter into a fixed-to-fixed swap. They exchange their principals, and
agree on the following conditions (for example): the American firm pays the British firm
4.75% on the GBP principal, while the British firm pays the American firm 6.50% on the
USD principal.
Lender Lender
(Britain) (United States)
4.50% 6.75%
GBP USD
6.50% USD
British firm American firm
4.75% GBP
Step 3: For the interest payments, the British firm has to pay 4.50% for its GBP loan. It
receives 4.75% from the American firm on the same amount in GBP, and it has to pay 6.50%
back to the American firm on the USD principal. Thus, the British firm has reduced its USD
financing costs to 10:
British Firm' s cost = 6.50% USD + 4.50% GBP - 4.75% GBP
= 6.50% USD - 0.25% GBP
≈ 6.25% USD
10 In fact, USD and GBP interest rate differentials are not directly comparable, except as an approximation.
19.niapS
Had the British firm borrowed USD directly, its cost would have been 6.50%. Similarly, for
the American firm, we have:
American Firm' s cost = 4.75% GBP + 6.75% USD - 6.50% USD
= 4.75% GBP + 0.25% USD
≈ 5.00% GBP
Had the American firm borrowed GBP directly, its cost would have been 5.25%. Thus, both
firms have saved approximately 25 basis points each year for five years.
Step 4: After five years, the two companies exchange again their principals, which are used to
repay their respective loans.
Suppose that firm A needs a fixed-rate EUR funding, while bank B needs floating-rate USD
funding, both for a period of five years. The market conditions for both firms are the
following:
It is clear that the bank has an absolute borrowing advantage in both EUR and USD. But the
firm has a comparative advantage in the USD market. Since the firm wants to borrow on the
market where it has a comparative disadvantage, a swap could be arranged as follows:
Step 1: The firm A borrows in USD at 6-M LIBOR + 0.50%, while bank B borrows in EUR
at a fixed 4% rate.
Step 2: Firm A and bank B enter into a fixed-for-floating swap. They exchange their
principals, and agree on the following conditions (for example): the firm pays the bank a fixed
4% interest rate on the EUR principal, while the bank pays the firm a floating LIBOR –
0.25% on the USD principal.
Lender Lender
(Europe) (United States)
4% EUR
20.niapS
Step 3: For the interest payments, the firm has to pay the 6-month USD LIBOR + 0.50% for
its USD loan. It receives the 6-month USD LIBOR – 0.25% from the bank on the principal
amount in USD, and it has to pay 4% back to the bank on the EUR principal. Thus, the firm
has reduced its EUR financing costs to 11:
Firm' s cost = (LIBOR USD + 0.50% USD) - (LIBOR USD - 0.25% USD) + 4% EUR
= 4 % EUR + 0.75% USD
≈ 4.75% EUR
Had the firm borrowed EUR directly, its cost would have been 5%. Similarly, for the bank,
we have:
Bank' s cost = 4% EUR + (LIBOR USD - 0.25% USD) - 4% EUR
= LIBOR USD − 0.25% USD
Had the bank borrowed USD directly, its cost would have been LIBOR. Thus, both firms
have saved approximately 25 basis points each year for five years.
Step 4: After five years, the two counterparties exchange their principals again, which are
used to repay their respective loans.
11 In fact, USD and EUR interest rate differentials are not directly comparable, except as an approximation.
21.niapS
Floating Rate
Lender
Similarly, if a firm treasurer believes short-term interest rates will decrease, he can use a swap
to convert an existing fixed-rate debt into a synthetic floating-rate liability. In the swap, the
firm will pay a floating rate and receive a fixed rate, which will result in lower interest cost if
the short-term interest rates decrease.
Fixed Rate
Lender
Fixed rate
Fixed rate
Swap Firm
Counterparty
6-M USD LIBOR
Consider the following example: a French firm has issued on October 1 a 5-year fixed-rate
bond for an amount of 10 Mio. USD. The exchange rate EUR/USD was 1.17 at the time of
issue. After one year, the EUR/USD exchange rate has increased to 1.25. The firm is worried
of a sudden rise of the USD in the next four years and decides to enter into a currency swap to
hedge this risk. The French firm will enter into a currency swap with a bank; the firm will pay
interests in EUR on an amount of 8 Mio. EUR and will receive interest in USD on an amount
of 10 Mio. USD. Of course, the firm should select to enter into a swap in which it will receive
a USD interest rate as close as possible to the USD interest rate it has to pay on its loan.
22.niapS
Lender
(United States)
interests in USD
interests in USD
Firm
Bank B
interest in EUR
Thus, through the swap, the firm is protected against a sudden appreciation of the USD (if
any!) and has locked in the profit from the USD decrease.
Swaps can also be used to lock in gains and stop losses on fixed-rate investments. For
example, a fixed-rate portfolio manager, after a strong decrease of the interest rates, can lock-
in the capital gain from his portfolio by swapping it against a floating-rate investment.
Last but not least, swaps can also be used in order to improve portfolio performance, on
both fixed-rate and floating-rate investments. Consider the following example: an investor has
a portfolio of floating-rate securities (USD 6-M LIBOR). To increase the yield on this
portfolio, he can enter simultaneously into two swaps: in the first one (the “original”), he will
swap his floating rate (USD 6-M LIBOR) against a fixed rate, let's say 10%; in the second one
(the “reverse”), he will swap a lower fixed rate (let's say 9.50%) against the same floating rate
(USD 6-M LIBOR). After the two transactions, the new rate of return on the portfolio is USD
6-M LIBOR, plus 50 basis points.
23.niapS
Floating rate
portfolio Swap
Counterparty 1
10%
Swap Investor
Counterparty 2
6-M USD LIBOR
Through swaps, it is also possible to create synthetic foreign currency assets, or to lock in
gains or stop losses on foreign currency investments.
Note that swap futures have also been introduced on a variety of exchanges. These contracts
are primarily designed to hedge cash market interest rate swaps.
24.niapS
• an interest rate swap generally involves neither an initial investment, nor a final maturity
payment as only coupon flows are swapped. This is not problematic if both parties
simultaneously buy and sell the same equivalent amount of a floating rate and a fixed rate
security, as this creates a netting of the two cash flows both at commencement and at
maturity.
• the expected cash flows under the floating rate security are uncertain, as they are
contingent on the level of future interest rates. However, this is a well know issue for all
floating rate notes. We will only examine the simplest solution hereafter.
Let us assume that under the terms of a swap, a financial institution receives fixed-rate
payments and makes floating-rate payments at the same time. Let us denote by V the value of
the swap, B1 the value of the fixed-rate bond underlying the swap, B2 the value of the
floating-rate bond underlying the swap, Q the notional principal of the swap agreement, R0,t
the effective discount rate corresponding to maturity ti, and K the fixed payment
corresponding to the fixed interest to be paid at time ti. At any time, the swap value may be
expressed as
V = B1 – B2
Since B1 is the present value of the cash flows from the fixed-rate bond, we have
n
K Q
B1 = ∑ +
i =1 (1 + R 0, t i ) ti
(1 + R 0, t n ) t n
For bond B2, the expected cash flows are not certain, as they are contingent on the level of
future interest rates. But in practice, bond B2 will have a value close to par, because of its
frequent re-pricing characteristics 12. Furthermore, when entering in the swap and immediately
after a coupon rate reset date, the value of bond B2 is equal to the notional amount Q. Using
this fact, we may write
12 As bond B2 coupon rate is frequently readjusted to a market rate, its yield to maturity is regularly equal to the
market required yield, and thus, after a readjustment (unless the credit quality deteriorates) its price is reset at
par.
25.niapS
K* Q
B2 = +
(1 + R 0, t1 ) t1
(1 + R 0, t1 ) t1
where K* is the floating amount used for the payment at date t1 (initially known).
Note that in the above, we have used so far effective discount rates. In practice, one might
have to use discount rates quoted according to a different compounding convention. For
instance, if we use continuous time finance, and denote the continuously-compounded
discount rate corresponding to maturity ti by r0, t i , the value of our two bonds are given by
n − r ⋅t
B1 = ∑ K ⋅ e 0, t i i + Q ⋅ e 0, t n n
− r ⋅t
i =1
and
− r0, t1 ⋅t1 − r0, t1 ⋅t1
B2 = K * ⋅ e + Q⋅e
At the beginning of the swap, the swap value should be zero, as the swap transaction is
ordinarily arranged at current market rates in order for the net present value of payments to
equal zero. That is, the fixed rate on a typical interest rate swap is set so that the market value
of the net floating rate payments exactly equal the market value of the net fixed-rate
payments. If the swap is not arranged as a zero-net-present-value exchange, one party pays to
the other an amount equal to the difference in the payments' net present value when the swap
is arranged.
At the end of its life, the swap value is also zero. During its life, the swap may have a positive
or negative value, depending on the value of B1 and B2, that is, depending on the interest rates
variations.
Example:
A company has entered in an interest rate swap, in which it has agreed to pay 6-month LIBOR and
to receive a 4% fixed rate (with semi-annual compounding) on a notional principal of
100 Mio. USD. The swap has a remaining life of 1.25 years. The relevant discount rates with
continuous compounding for 3-month, 9-month, and 15-month maturities are 5%, 5.5%, and 6%
respectively. The LIBOR rate at the last payment date was 5.10% (with semi-annual
compounding). What is the value of the swap?
We have K = 2 Mio. USD, and K* = 2.55 Mio. USD. The fixed bond price is:
n − r0, t i ⋅ t i
∑ K ⋅ e
− r0, t n ⋅ t n
B1 = + Q⋅e
i =1
− 0.25⋅0.05 − 0.75⋅0.055 −1.25⋅0.06 −1.25⋅0.06
= 2⋅e + 2⋅e + 2⋅e + 100 ⋅ e
= 98.52 Mio. USD
and the floating bond price is:
− r0, t1 t 1 − r0, t1 t 1 −0.25⋅ 0.05 −0.25⋅ 0.05
B 2 = K * ⋅e + Q⋅e = 2.55 ⋅ e + 100 ⋅ e
= 101.28 Mio. USD
Hence, the value of the swap for the counterparty that pays a floating rate and receives a fixed rate
is V = 98.52 – 101.28 = –2.75 Mio. USD. For the other counterparty, which pays a fixed interest
and receives a floating rate, the value of the swap is +2.75 Mio. USD.
26.niapS
The payment to the company at each payment date (each six months) is:
⋅ (Floatingrate − Fixedrate)
Q
2
where the floating rate is the rate that prevailed at the previous payment date. This is a normal
forward contract on the floating rate, except that it is always settled six months in arrears.
Suppose now that Fi is the forward interest rate (expressed with semi-annual compounding)
for the 6-month period prior to the payment date i (i > 1). The value of the forward contract
for the payment number i (i > 1) for the party receiving a fixed-rate payment (amount: K) and
paying a floating-rate one (amount 0.5 ⋅ Fi ⋅ Q) is therefore
1 1
K − ⋅ Fi ⋅ Q ⋅
(1 + R 0, t i ) i
t
2
At the first payment date, we have a payment of K* and a receipt of K. The actual value of the
first payment is therefore
(K − K )⋅ (1 + R1 *
) t1
0, t1
The total value of the swap for the party that receives fixed and pays floating is therefore (if
there are n semi-annual payments):
( ) (1 + R1
n
1 1
V = K − K* ⋅ t1
+ ∑ K − ⋅ F ⋅i Q ⋅
i=2 (1 + R 0, t i ) i
t
0, t1 ) 2
For the party that receives floating and pays fixed, the value of the swap is:
( ) (1 + R1 1
n
1
V' = K* − K ⋅ t1
+ ∑ ⋅ Fi ⋅ Q − K ⋅
i=2 2 (1 + R 0, t i ) i
t
0, t1 )
Note that using continuous time relationship, and denoting by r0, t i the continuously
compounded discount rate corresponding to maturity ti, one would get:
( ) − r ⋅t
n
1
+ ∑ K − ⋅ Fi ⋅ Q ⋅ e 0 ,i i
− r0 ,1 ⋅ t 1
V = K − K* ⋅ e
i=2 2
and
( ) 1 − r ⋅t
n
+ ∑ ⋅ Fi ⋅ Q − K ⋅ e 0 ,i i
− r0 ,1 ⋅ t 1
V' = K* − K ⋅ e
i=2 2
Let us illustrate this with a numerical example:
27.niapS
Example:
A company has entered in an interest rate swap, in which it has agreed to pay 6-month LIBOR and
to receive a 4% fixed rate (with semi-annual compounding) on a notional principal of
100 Mio. USD. The swap has a remaining life of 1.25 years. The relevant discount rates with
continuous compounding for 3-month, 9-month, and 15-month maturities are 5%, 5.5%, and 6%
respectively. The LIBOR rate at the last payment date was 5.10% (with semi-annual
compounding). What is the value of the swap?
The first payment will be 2 Mio. USD (fixed leg) against 2.55 Mio USD (floating leg, based on the
LIBOR rate at the last payment date).
For the next two payments, we need to determine the relevant forward rates. Let us denote
r0,0.25 = 5%, r0,0.75 = 5.5%, r0,1.25 = 6%. Since our spot interest rates are continuously compounded
and annualized, we have:
Using the same methodology, we can set the fixed rate of a fixed against floating interest rate
swap so that its initial value is zero. Let us illustrate this by another example.
Example:
A company wants to pay a floating 6-month LIBOR and receive a fixed rate (annual payment) in a
two-year CHF interest rate swap for a notional amount of 100 Mio. CHF. Today’s zero-coupon
rates are as follows: R0, 0.5 = 6.50% p.a., R0,1 = 7.00% p.a., R0,1.5 = 7.25% p.a., R0, 2 = 7.35% p.a.
These rates are effective interest rates. What should be the fixed rate on such an interest rate swap
with annual payments?
First, let us determine the forward rates from the spot rates. We will use simple and non-
annualized interest rates in order to illustrate the process in such a case.
• For the 6-month contract, the forward rate equals the spot rate:
F0,0.5 = R 0,0.5 = (1.0650) − 1 = 3.1988%
0.5
• For the 6-month forward rate in 6 months, go long the 1 year (7%), and short the half year
(6.5%):
1.07
F0.5,1 = − 1 = 3.6834%
(1.0650)0.5
• For the 6-month forward rate in 12 months, go long the 1.5 year (7.25%), and short the 1 year
(7%):
F1,1.5 =
(1.0725)1.5 − 1 = 3.8035%
1.07
• For the 6-month forward rate in 18 months, go long the 2 years (7.35%), and short the 1.5 year
(7.25%):
28.niapS
F1.5, 2 =
(1.0735)2 − 1 = 3.7548%
(1.0725)1.5
Now, determine the present value of the expected (i.e. implied) floating payments.
1 1 13.2247
13.2247 = R ⋅
1.07 + 1.07352 ⇒ R = 1.8023 = 7.34% p.a.
Thus, 7.34% is the arbitrage-free fixed rate in this two-year swap.
Example 13:
A financial institution has entered in a swap in which it receives a fixed rate of 5% in JPY and
pays a fixed rate of 8% in USD. All payments are made annually; the principals in the two
currencies are 10 Mio. USD and 1'200 Mio. JPY. The swap has a remaining life of three years, the
Japanese interest rate is 4% p.a., and the USD interest rate is 9% p.a. (both term structures are
supposed to be flat, and these rates are given with continuous compounding). What is the value of
such a JPY/USD swap, if the current USD/ JPY exchange rate is 110?
The domestic bond is an 8% USD coupon-paying bond on a principal of 10 Mio. USD. Its value
is:
−0.09 −0.09 ⋅ 2 −0.09 ⋅3
BD = 0.8 ⋅ e + 0.8 ⋅ e + 10.8 ⋅ e = 9.64 Mio.USD
The foreign bond is a 5% JPY coupon-paying bond on a principal of 1'200 Mio. JPY. Its value is:
−0.04 −0.04 ⋅ 2 −0.04 ⋅3
BF = 60 ⋅ e + 60 ⋅ e + 1260 ⋅ e = 1'230.55 Mio. JPY
13 Source: HULL John C., 2005, “Options, Futures, and other Derivatives”, 6th edition, Prentice Hall
International, New Jersey.
29.niapS
1230.55
The value of the swap is − 9.64 = 1.55 Mio. USD
110
For the counterparty that pays JPY and receives USD, the value of the swap is –1.55 Mio. USD.
Thus, using the term structure of forward exchange rates and the term structure of domestic
interest rates, the value of a currency swap can be determined.
Example 14:
A financial institution has entered in a swap in which it receives a fixed rate of 5% in JPY and
pays a fixed rate of 8% in USD. All payments are made annually; the principals in the two
currencies are 10 Mio. USD and 1'200 Mio. JPY. The swap has a remaining life of three years, the
JPY interest rate is 4% p.a., and the USD interest rate is 9% p.a. (both term structures are supposed
to be flat, and these rates are given with continuous compounding). What is the value of such a
JPY/USD swap, if the current USD/ JPY exchange rate is 110?
The current spot rate is 110 JPY/USD, or 0.009091 USD/JPY. The interest rate difference between
USD and JPY is 5% p.a. We can derive the 1-year, 2-year, and 3-year forward exchange rates as
0.05⋅1
F0,1 = 0.009091 ⋅ e = 0.0096
0.05⋅2
F0, 2 = 0.009091 ⋅ e = 0.01
0.05⋅3
F0,3 = 0.009091 ⋅ e = 0.0106
respectively. The exchange of interest payments involves receiving 60 Mio. JPY and paying
0.8 Mio. USD. The risk-free rate in USD is 9% p.a. The value of the forward contracts
corresponding to the exchange of interest rates is therefore (in Mio. USD):
−0.09⋅1
(60 ⋅ 0.0096 − 0.8)⋅ e = −0.21
−0.09⋅2
(60 ⋅ 0.001 − 0.8) ⋅ e = −0.16
−0.09⋅3
(60 ⋅ 0.0106 − 0.8) ⋅ e = −0.13
The final exchange involves receiving 1’200 Mio. JPY, and paying 10 Mio. USD. The value of the
forward contracts corresponding to this is therefore (in Mio. USD):
−0.09⋅3
(1'200 ⋅ 0.0106 − 10) ⋅ e = 2.04
Thus, the total value of the swap is 2.04 – 0.13 – 0.16 – 0.21 = 1.54 Mio. USD, which is in
agreement with the results of our previous example (allowing for rounding errors).
30.niapS
The simplest case to consider is the example of a well-diversified (highly correlated with the
S&P 500) portfolio owner who wants to hedge against downside market risk for a given
amount of time. A possible solution is to enter into a swap agreement with a swap dealer to
exchange the S&P 500 total return (including capital appreciation/depreciation and dividends)
against a floating interest rate.
The simplest case to consider is the example of a commodity producer, who wants to hedge
against commodity price variations. He can enter into a commodity swap, in which he will
pay to the swap dealer a per unit floating price (for instance based on the average market spot
price), and receive a fixed determined price. As he produces one unit of the commodity, he
will sell it on the spot market, and receive the corresponding per unit spot price, that he will
use to pay the swap dealer.
Through this swap, the commodity producer has the insurance to receive a fixed
predetermined price for its commodity. No exchange of notional take place, all transactions
are cash-settled.
31.niapS
average commodity
spot price spot price
Commodity
Spot SWAP
producer
Market commodity fixed price DEALER
Note that the commodity swap can also be used in order to speculate, without the transactions
on the spot market.
Example:
Consider a volatility swap with a notional amount of EUR 100,000 per volatility percentage point
and a delivery price of 20 percent. If, at maturity, the annualized realized volatility over the
lifetime of the contract settled at 21.5 percent then the buyer of the swap would receive EUR
150,000. If, at maturity, the annualized realized volatility over the lifetime of the contract settled at
18.5 percent then the seller of the swap would receive EUR 150,000.
Similarly, a variance swap is a forward contract on realized variance, which is the square of
the realized volatility.
2
N −1 S
1
VR = σ = 252 ⋅
2
⋅ ∑ Log t i+1
R
N − 2 i =1 St
i
where St i is the closing level of the stock at time ti, and N is the number of business days from
the trade date up to and including the maturity date. Its payoff at expiration is equal to:
=Payoff Notional Amount ⋅ ( Realized Variance – Variance Strike Price )
The notional amount is usually expressed in terms of monetary amount per volatility point
squared. When entering the swap the Variance Strike Price is typically set at a level so that
the counterparties do not have to exchange any cash flows (‘fair strike’).
32.niapS
A German firm uses 25'000 barrels of oil every three months [a barrel is a common unit of
measurement for the daily volume of crude oil produced by a well or from a field. The
volume of a barrel is equivalent to 42 US gallons, or 158.99 litres]. On the international
markets, oil is priced in USD; thus, the German firm is exposed to both oil price variations
and to the EUR/USD exchange rate fluctuations. The German firm would like to engineer a
structure that would allow it to fix the price of oil in EUR.
The current spot EUR/USD exchange rate is 1.25. Seven-year oil swaps are available at
70 USD a barrel, seven-year USD interest swaps are available at a 5% rate against 3-month
LIBOR, and seven-year EUR-for-USD currency swaps are available at a rate of 4% against 3-
month LIBOR.
Step 1: Enter into a commodity swap to receive the average oil spot price and pay a fixed
price. The fixed-price leg of the swap is given by 25'000 barrels times 70 USD, i.e. a total of
1'750'000 USD.
Step 2: Enter into an interest rate swap to received a fixed rate and pay a floating rate, both in
USD.
The notional principal required on a USD interest rate swap for the fixed side to generate
1'750'000 USD every three months is given by dividing the quarterly cash flow requirements
by the periodic rate (5% p.a., or 1.25% quarterly).
1'750'000
=140'000'000 USD
0.0125
Step 3: Enter into a currency swap to receive USD and pay EUR.
The present value of the 28 periods cash flows on the fixed-rate side of the interest rate swap
at a 5% annual rate is approximately 41'256'356 USD. At the current spot rate of EUR/USD
1.25, this represents a present value in EUR of 33'005'085 EUR. This present value
corresponds to 28 periodic payments of 1'354'588 EUR, using the current EUR rate of 4%
(1% quarterly). From there, we can determine the notional principal that would generate the
quarterly payments of 1'354'588 EUR at the 4% rate; this is 135'458'800 EUR.
33.niapS
Spot oil
market
Commodity
spot price 1'750'000 USD Swap
oil
Dealer
average spot
price
1'354'588 EUR
The final structure is represented above. It gives the German firm a fixed price of 54.18 EUR
per barrel for the next seven years (1'354'588 EUR / 25’000).
34.niapS
2. Credit derivatives*
The credit derivatives market growth since the mid 1990’s has been spectacular, as illustrated
in the following figure:
Figure 2-1: The notional amount of credit derivatives globally is larger than the global
amount of debt outstanding
According to the International Swaps and Derivatives Association estimates, the notional
amount of credit derivative contracts outstanding in mid 2008 was USD 54.6 trillion, i.e. a 10
times growth since 2001. One can attribute this tremendous growth in the credit derivatives
markets to standardization of documentation, a more diverse base of market players and
innovations in terms of products. Credit derivatives have become main stream and are
increasingly used in diverse investment activities. However, this growth has also resulted in
serious concerns. First, the notional value of credit derivatives contracts dramatically exceeds
the notional value of underlying credit instruments, which could trigger a market squeeze in
case of physical delivery. As an illustration, when Delphi filed for bankruptcy on October 8
2005, there was approximately USD 2 billion of bonds trading at around 60% of their face
value, but an estimated USD 25 billion in credit derivatives linked to this debt. Second, there
is still an important backlog of “waiting to be confirmed” contracts, despite the Federal
Reserve request to act in 2005.
35.niapS
Credit default swaps on single issuers represent the majority share of credit derivatives,
although market participants are increasingly using index products on baskets of credit default
swaps.
36.niapS
2.2.1 Definition*
A credit default swap is essentially a bilateral agreement in which one counterparty (the
protection buyer) pays a periodic premium to another counterparty (the protection seller) in
return for a contingent payment if and only if a predefined credit event occurs in a pre-
specified reference bond, such as bankruptcy or debt restructuring. This contingent payment
will typically compensate the protection buyer for the difference between par and the recovery
value of the reference bond following the credit event. The premium paid by the protection
buyer to the seller, often called “spread”, is quoted in basis points per annum of the contract’s
notional value and is usually paid quarterly.
Example:
The 5-year credit default swap for Ford was quoted around 160 bps p.a. on April 27, 2004. This
means that if you want to buy a 5-year protection for USD 10 million exposure to Ford credit, you
would pay 40 bps, or $40,000, every quarter as an insurance premium for the protection you
receive.
The mechanics of a CDS are fairly straightforward and is summarized in Figure 2-3. If no
default occurs, no payment is made on the CDS.
Note that:
• If the credit event happens, the buyer of protection stops paying the premium – he only
needs to pay the accrued premium up to the day of the credit event. This allows both
parties to close out their positions soon after the credit event and eliminates the ongoing
administrative costs that would otherwise occur.
• Although periodic payment dates are the most common, some swap contracts include an
up-front, lump sum payment.
37.niapS
From a terminology perspective, one can identify different types of CDS contracts:
• CDS: indicates that the underlying reference entities and obligations are senior unsecured
bonds, issued by corporate or sovereign issuers
• LCDS: Loan-only CDS refers to contracts where protection is bought and sold on
syndicated secured leveraged loans. These are higher in the capital structure and with
higher recovery rates than CDS.
• MCDS: The reference entity is a municipality and the reference obligation a municipal
bond.
• ABCDS: CDS on structured securities (Asset Backed Securities typically)
• Preferred CDS: CDS on Preferreds
From an economic perspective, all CDS are similar to buying insurance against credit events.
These contracts may therefore appear to be closer to options rather than swaps. Indeed, the
protection buyer has essentially the right to sell his bond at par if the credit event materializes.
However, with a CDS, the cost of the “option” is paid in instalments and not upfront as with a
traditional option. Note that CDS prices are normally quoted in basis points (bps) and are a
measure of the reference entity’s credit risk.
Therefore, in theory, the n-year CDS spread s should be close to the excess of the par yield of
an n-year corporate bond y over the par yield of an n-year risk free bond r:
s= y − r
otherwise arbitrage opportunities would exist.
This relationship holds only approximately for a number of reasons. In particular the arbitrage
argument assumes that:
• market participants can short corporate bonds. This is only partially true.
• market participants can borrow at the riskless rate. This is rarely the case.
• The argument does not take into account the "cheapest-to-deliver bond" option in a
credit default swap. Typically a protection seller can choose to deliver any of a number
of different bonds in the event of a default.
15 J.Hull, M.Predescu, A.White, The Relationship Between Credit Default Swap Spreads, Bond Yields, And
Credit Rating Announcements, Journal of Banking and Finance, 28 (Nov. 2004) pp 2789-2811
38.niapS
• The arbitrage assumes that interest rates are constant so that par yield bonds stay par
yield bonds. By defining the corporate bond used in the arbitrage as a par corporate
floating bond and the riskless bond as a par floating riskless bond the constant interest
rate assumption can be avoided. Unfortunately, in practice par corporate floating bonds
rarely trade.
• CDS bear counterparty default risk.
• Because of tax and liquidity reasons an investors could prefer a riskless bond to a
corporate bond plus a CDS or vice versa.
But the main problem with the above equation lies in choosing the risk-free rate, r. Bond
traders tend to regard the Treasury zero curve as the risk-free zero curve and measure a
corporate bond yield spread as the spread of the corporate bond yield over the yield on a
similar government bond. By contrast, derivatives traders tend to use the swap zero curve
(sometimes also called the LIBOR zero curve) as the risk-free zero curve in their pricing
models because they consider LIBOR/swap rates to correspond closely to their opportunity
cost of capital.
The choice of the Treasury zero curve as the risk-free zero curve is based on the argument that
the yields on bonds reflect their credit risk. A bond issued by a government in its own
currency has no credit risk so that its yield should equal the risk-free rate of interest.
However, there are many other factors such as liquidity, taxation, and regulation that can
affect the yield on a bond. For example, the yields on US Treasury bonds tend to be much
lower than the yields on other instruments that have zero or very low credit risk. One reason
for this is that Treasury bonds have to be used by financial institutions to fulfill a variety of
regulatory requirements. A second reason is that the amount of capital a financial institution is
required to hold to support an investment in Treasury bonds is substantially smaller than the
capital required to support a similar investment in low risk corporate bonds. A third reason is
that the interest on Treasury bonds is not taxed at the state level whereas the interest on other
fixed income investments is taxed at this level. For all of these non-credit-risk reasons, the
yields on U.S. Treasury bonds tend to be depressed relative to the yields on other low risk
bonds.
The swap zero curve is normally calculated from LIBOR deposit rates, Eurodollar futures,
and swap rates. The credit risk associated with the swap zero curve is somewhat deceptive.
The rates for maturities less than one year in the swap zero curve are LIBOR deposit rates and
are the short-term rates at which one financial institution is willing to lend funds to another
financial institution in the inter-bank market. The borrowing financial institution must have an
acceptable credit rating (usually Aa). From this it might be assumed that longer rates are also
the rates at which Aa-rated companies can borrow, but this is not the case. In fact the n-year
swap rate is lower than the n-year rate at which an Aa-rated financial institution borrows
when n > 1. It represents the credit risk in a series of short-term loans to Aa borrowers rather
than the credit risk in one long-term loan to Aa borrowers. Consider for example the 5-year
swap rate when 6 month LIBOR is swapped for a fixed rate of interest. This is the rate of
interest earned when a bank enters into the 5-year swap (receive fix-pay float) and makes a
series of 10 six-month loans to companies with each of companies being sufficiently
creditworthy that it qualifies for LIBOR funding at the beginning of its six-month borrowing
period. From this it is evident that rates calculated from the swap zero curve are very low risk
rates, but are not totally risk free. They are also liquid rates that are not subject to any special
tax treatment.
Empirical research shows that the credit default swap market appears to use the swap rate
rather than the Treasury rate as the risk-free rate. More precise calculations suggest that the
39.niapS
market is using a risk-free rate about 10 basis points less than the swap rate. Hence in the
above equation r=swap rate – 10 bps. This estimate is plausible, since 10 bps is a reasonable
default risk premium for a AA-rated 6-month instrument.
• A Bankruptcy refers to the reference entity’s insolvency or inability to repay its debt.
Bankruptcy is deemed to have occurred if it results in the default of the reference entity’s
obligations. The definition of bankruptcy is a written admission of a company’s inability to
pay its debt which must be made in a judicial, regulatory, or administrative filing.
• A Failure-to-Pay occurs when the reference entity, after a certain grace period, fails to
make payment of principal or interest when due. A minimum threshold amount is normally
nominated in the confirmation that must be exceeded before this event is triggered (by
default, USD 1 million).
• A Debt Restructuring refers to a change in the terms of debt obligations that is adverse to
creditors. Restructuring is by far the most problematic of these trigger events, because
“adverse change” may be an ambiguous concept. The main issue is that, unlike bankruptcy
or failure to pay, some restructuring of debt may not lead to losses for investors. Moreover,
even if investors suffer financial losses, the amount of losses is more difficult to determine,
if restructuring of debt involves an exchange of bonds with different coupons and/or
maturities. Accordingly, some market participants prefer to exclude the restructuring
provision from a credit derivative contract altogether, or to restrict the scope of the
provision. Currently, the ISDA agreement offers four options for treating the issue of debt
restructuring:
40.niapS
2.2.4 Settlement*
The first step taken after a credit event occurs is a delivery of a “Credit Event Notice,” either
by the protection buyer or the seller. The notice describes what exactly has occurred that the
triggering party believes constitutes a credit event. Once the credit event is verified, the Credit
Event Notice is accepted and the contract enters its settlement phase. The compensation is to
be paid by the protection seller to the buyer via either (1) physical settlement or (2) cash
settlement, as specified in the contract.
Physical settlement is the market standard. Following the credit event, it involves the
protection buyer delivering the notional amount of “deliverable obligations” to the protection
seller in return for the notional amount paid in cash.
In practice, the buyer can deliver any bond meeting certain criteria issued by the reference
entity that is pari passu (of the same level of seniority) as the specific bond referenced in the
contract. Usually a CDS specifies that a number of different bonds can be delivered in the
event of default. Those bonds do have the same seniority but they may not sell for the same
percentage of face value immediately after default. This creates a cheapest-to-deliver bond
option for the holder of the CDS. In the case of default, the buyer of the protection will
purchase the cheapest bond on the market that fulfils the delivery requirements. The reasons
for this phenomenon are: the claim that is made in the event of default is typically equal to the
bond’s face value plus accrued interest. Thus bonds with high accrued interests at the default
point in time tend to have higher prices right after default. Also, the market might assume that
in the event of a reorganization of the company some bond holders will do better than others.
Unless already holding the deliverable asset, the protection buyer may prefer a cash
settlement in order to avoid any potential squeeze on the underlying asset. Cash settlement
will also be the choice of a protection buyer who is simply using a default swap to create a
synthetic short position in a credit.
Given the nature of a bilateral agreement between two parties they can agree to unwind the
trade based on the market price of the defaulted bond (for example 40 for 100) which usually
would be the average of the prices of a certain number of market makers for this bond. The
seller then pays the net amount (100 – 40) = 60. This is called ‘cash settlement’. The recovery
value of the bond (in the example 40) is not fixed and can only be determined after the credit
event.
41.niapS
Example:
Suppose an investor owns 10 million bonds of 6.75%, February 7 2010 Ford car company. He is
afraid that the credit quality will deteriorate and buys a three year protection maturing on March
20 2010. The quoted spread is 167 bps at the time of the contract. As can be seen in the following
CDS cash flow graphs, the investor has to pay 10’000’000 ∙ 0.25 ∙ 0.0167 = 41’750 per quarter to
the seller of this protection as long as no default occurs. In actuality, one would calculate the
number of effective days (A) between the quarters leading to (A/360) ∙ 10’000’000 ∙ 0.0167. In the
case of default, the premium due from the protection buyer accrues until the date of default and the
seller of protection has to deliver the contingent payment.
No default
Default at time t
$ 41‘750 $ 41‘750 $ 41‘750 $ 41‘750
Deliver asset in exchange for
$10m
Net value is $ 6 m
Default time
Firstly, one has to look up existing CDS contracts (or set-up new ones). Let’s look for
examples of outstanding CDS in the telecom industry. For the purpose of the example, let us
pick the credit default swap of Deutsche Telekom AG. It is a CDS with a maturity of 5 years
trading at a spread of 24 bps. This spread has narrowed 8.295 bps in the last three month or
13.92 bps since of the start of the year.
42.niapS
Figure 2-5: Finding a Credit Default Swaps from a list of existing CDS
To look at our chosen CDS in more detail with the help of the Bloomberg screen ‘CDSW’.
As the print screen shows we have a number of parameters on this CDS on Deutsche
Telekom. The information is composed of (i) details on the deal such as the issuer, nominal
amount of 10 million EUR, effective date April 07, maturity date April 12, frequency of
payments quarterly, assumption on the recovery value of 40 etc.); (ii) a convenient calculator,
which sets the spread of the deal to 24.899 bps per cash settlement date; and (iii) information
on where CDS of different maturities are trading in the market.
43.niapS
Sub-indices Financials (24) Financials (15) Financials (10) Korea (8) none none
Consumer goods Autos (10) Investment Greater
(34) Consumer goods goods (10) China (9)6
Energy (15) cyclical (15) Technology (10) Rest of
Industrial (30) Consumer goods non- High volatility Asia (13)7
TMT (22) cyclical (15) (10)
High volatility (30) Energy (20)
B (44) Industrial (20)
BB (43) TMT (20)
HB (30) High volatility (30)
This table represents the newest series, DJ CDX and DJ iTraxx. The number of reference entities is in parenthesis. 2
Maximum number per sector is 10. 3) Only countries: Brazil, Bulgaria, Columbia, Korea, Malaysia, Mexico, Panama, Peru,
Philippines, Rumania, Russia, South Africa, Turkey, Venezuela 4) Largest and most liquid non-financials of iBoxx-EUR-
Corporate bond index. 5) Most liquid entities outside of the financial index with ratings BBB/Baa3 or lower and negative
outlook. 6) China, Hong Kong, Taiwan with at least 2 entities. 7) India, Malaysia, Philippines, Singapore and Thailand
These CDS indices typically have a fixed composition and fixed maturities specific to the
products. A particular name remains in the index until the CDS is triggered due to a credit
event. New indices – called on-the-run-indices – are being created periodically (usually twice
a year) with new underlying credits. There are strict rules which define the selection process
for single-name CDS to be included in these indices. Credits that are prominent in the market
and have liquid single-name credit default swaps are prime candidates for inclusion. Some of
the indices have sub-indices based on ratings, industry groups or geographic region.
Considerations of duration and liquidity are also part of the rationale to create new indices.
The existing indices – called off-the-run – continue to trade until maturity. Of course, the on-
the-runs tend to be more liquid than the off-the-runs.
44.niapS
For example, consider the iTraxx® Europe index. This index comprises 125 investment grade
rated European entities [10 Autos, 30 Consumers, 20 Energy, 20 Industrials, 20 TMT-
Telecommunications Media and Technology, 25 Financials] with the highest CDS trading
volume, as measured over the 6 months before the launch date. Every March 20th and
September 20th (roll dates) a new series is launched. The new launched index start on the roll
date. Entities are weighted equally. Before launch of a new index series conference calls are
organized between the market makers to agree on index composition, reference obligations,
coupon level and recovery rate.
CDS references the credit spread at launch (“premium”) of the most current series. In the case
of the iTraxx series 9 Europe 5-year exposure the premium is 165 bps.
Assume that three days after launch, the market is 145 bps (i.e. spreads have shrunk) and an
investor wants to take EUR 10 Mio iTraxx Europe exposure in unfunded/CDS form by
becoming protection seller. CDS is executed at the premium level. The market maker pays
165 bps per annum quarterly to the investor on a notional amount of EUR 10 Mio. The
present value of the difference between premium (165 bps) and fair value (145 bps) of the
CDS is settled through an upfront payment on T + 3 days. The investor pays the present
value of 20 bps plus accrued interest to market maker (EUR 90,450). The present value is
typically calculated via the CDSW function on Bloomberg.
If during the lifetime of the contract no Credit Event occurs, the investor will continue to
receive premium (165 bps) on the original notional amount until maturity.
45.niapS
At the contrary, if a Credit Event occurs (e.g. a Credit Event occurs on a Reference Entity, for
example, in year 3, and the Reference Entity weighting is 0.8%), the investor pays to market
maker (0.8% · 10'000'000) = 80'000 EUR, and the market maker delivers to the investor EUR
80'000 nominal face value of Deliverable Obligations of the Reference Entity. The notional
amount on which premium is paid reduces by 0.8% to 99.2% x 10'000'000 =EUR 9'920'000,
and after the Credit Event, the investor receives premium of 165 bps on EUR 9.92 Mio until
maturity subject to any further credit events.
The following Figure shows a partial list of index members with their according weights for
the iTraxx series 9 Europe 5-year index. The last column titled 'Spread' gives the latest spread
value for the corresponding five year CDS ticker.
Figure 2-8: Partial list of index members with their respective weightings
CDS indexes are important innovations that allow financial players to trade a broader
spectrum of credits at lower cost and in a more liquid market. There are now tradable CDS
indexes that cover specific sectors, points on the credit curve, geographies, and credit
qualities. In addition, banks offer CDS index tranches and exchanges have introduced CDS
index options to allow speculators to tailor their exposure to credit risk. Trading CDX
tranches allows traders to take positions on correlation between credit events within a CDX
index or across different CDX indexes, and CDX options allow speculators to take a position
on the volatility of CDS market spreads.
46.niapS
Broadly defined, a CLN is a debt instrument with principal contingent upon the credit
performance of a specific reference asset issued by a reference entity. Provided the reference
entity experiences no credit event during the life of the CLN, the principal will be repaid to
the investor on maturity. Should the reference entity experience a credit event, the CLN is
redeemed. But instead of receiving the principal amount originally invested, the CLN investor
receives a bond issued by the reference entity. CLN investors therefore bear the potential
losses due to default of the reference asset. In return, they pick up some additional yield,
usually under the form of a higher coupon.
How does this work in practice? Say for instance that a bank issues medium term notes and
wants to structure a CLN. Typically:
• The bank selects a reference entity and sells protection using a CDS on that selected
reference entity. Selling protection means the bank receives a regular fixed payment from
the CDS counterparty.
• The bank issues the CLN for the same principal amount and maturity as the CDS. The final
terms of the CLN should mirror the terms in the CDS transaction. The CLN investor pays
the bank to buy the note.
• The bank has to pay the investor regular interest until the maturity of the CLN.
Figure 2-9: The bank sells credit protection and issues the CLN. Investors buy the CLN
and receive regular coupons
Provided there is no credit event by the reference entity, the investor will receives back the
principal investment on the maturity of the CLN, like in a straight bond position.
47.niapS
But if the reference entity experiences a credit event, the investor will experience a credit loss,
as
• The CDS on which the bank sold protection is triggered. The bank has to pay to the CDS
counterparty the principal amount of the CDS in cash. The bank receives in return a
deliverable instrument – a bond that was issued by the reference entity that is now in
default.
• The CLN is also triggered. The investor does not get his principal returned, instead the
bank delivers the bond to the CLN buyer. The investor will have experienced a loss as a
result of the credit event because the delivered bond will be worth less than the original
sum invested. The scale of the loss incurred will depend on the market value of the
delivered bond.
In essence, the position of CLN investors is the same as if they had sold default protection on
the reference asset via a CDS to the CLN issuer. That allows them to obtain a higher coupon
on the CLN than would have been achieved on a normal bond.
Example
An investor has USD 10 Mio to invest for 10 years in Mexico risk. The investor can buy USD 10
Mio of a 10-year Mexican bond with a 5.75% yield. Alternatively, the investor can put USD 10
Mio in a 10-year risk-free deposit (yield: 4.95%), and sell a CDS on USD 10 Mio of 10-year
Mexican debt, for an annual premium of 1.2%. The risk-free rate (4.95%) plus the premium
(1.2%) is higher than the yield on the bond (5.75%). In this latter case, the investor has
synthetically re-created a CLN.
Note that
• The holder of a CLN has some credit exposure to the issuer of the notes as well as to the
reference entity or entities as defined in the terms and conditions of the CLN.
• Investors who are restricted from using credit derivatives because of operational, legal or
regulatory constraints can still create the investments they want using a CLN.
• CLNs can easily be tailored to suit investors needs in terms of coupons, maturity,
underlying, etc.
• CLN can be exchange-listed and rated by a rating agency.
CLNs became increasingly popular from 1997 as banks needed to reduce their exposure to
risk of default in loans to emerging markets or companies. More recently, they also helped
banks limiting their regulatory capital requirements as CLNs do not represent claims against
the underlying borrower.
CLNs may be Single Name, Basket or Nth-to-Default. Let us briefly review what these terms
mean.
48.niapS
A Single Name CLN has only one reference entity. If no credit event occurs during the
lifetime of the CLN, the CLN holder regularly receives a fixed coupon and receives his
principal back at the maturity of the CLN. If a credit event occurs prior to maturity date of the
CLN, no further coupon payments are paid and the single name CLN is redeemed by
delivering the reference asset.
A Basket CLN references a basket of reference entities. Investors will receive a fixed pre-
specified coupon until the earlier of the maturity date of the CLN or the date on which credit
events have occurred in respect of every reference entity. In a typical basket CLN, if a credit
event occurs the nominal amount of the CLN is reduced by the same proportion as the
relevant reference entity bears to the basket and the investor will be paid an amount equal to
the recovery value of outstanding obligations issued by the relevant reference entity (cash
redemption). Thereafter, as the nominal amount of the CLN has been reduced, coupon
payments are reduced proportionally and/or the coupon rate may be reset.
Note that unlike with a single name CLN, a Basket CLN is not terminated following a credit
event. The remaining nominal capital continues to be exposed to potential credit events
throughout the remaining term. With a basket CLN, the loss potential per reference entity is
therefore limited to the proportional weighting of each reference entity within the basket. For
instance, if there are five equally weighted entities in the Basket CLN, an investor can lose a
maximum of 20% (minus recovery value) of the nominal capital per credit event.
Example
Consider a EUR 10 Mio basket CLN issued by a bank where the reference portfolio is made of 5
allocations of EUR 2 Mio to bonds A, B, C, D, and E. The CLN is issued at par
During the life of the CLN, if there is no default, the cash flows are as follows (the coupon can be
fixed or floating):
Coupon on EUR 10 Mio
CLN Issuer CLN
(Bank) Investor
5×EUR 2 Mio credit protection
If reference entity B defaults, EUR 2 Mio of the basket CLN are redeemed and the cash flows are
as follows:
Coupon on EUR 8 Mio
CLN Issuer CLN
(Bank) 4×EUR 2 Mio credit protection Investor
49.niapS
An Nth-to-Default CLN is a note which references a basket of reference entities. Upon the
occurrence of a predetermined number (N) of credit events in relation to the reference entities,
the entire Nth-to-Default CLN terminates – whereas a Basket CLN would continue with a
reduced nominal amount. A first-to-default CLN terminates after one reference entity suffers
a credit event, a second-to-default after two reference entities would suffer a credit event, etc.
In contrast to a Basket CLN, with an Nth-to-Default CLN the loss potential per reference
entity is not limited to the proportional weighting of each reference entity within the basket
but instead to the entire nominal amount of the Nth-to-Default CLN. When a credit event
occurs in respect to the Nth reference entity in the basket, no further coupon payments are
paid to the investor and the Nth-to-Default CLN is redeemed by payment to the investor of an
amount equal to the market value of obligations issued by the relevant reference entity (cash
redemption). In other words, in the case of an Nth-to-Default CLN, the investor could lose a
maximum of 100% (and not just 20%) of the nominal amount when one reference entity
experiences a credit event even if there are five entities in the Nth-to-Default CLN.
50.niapS
Tranched credit structures increase market efficiency by reallocating the risk of credit losses
on an underlying portfolio into tailored investments that satisfy an investor’s desired
risk/return profile. Synthetic CDOs have historically been customized in nature, meaning that
the end investor selects the underlying portfolio, amount of subordination, and tranche size.
Currently, there are standardized synthetic collateralized debt obligations using the credits in
the Dow Jones CDX and iTraxx indices. These tranched index products create a standardized,
liquid, and transparent instrument to trade tranched credit risk. It also introduces the ability to
hedge other CDO structures or take a view on market implied correlation, a key value driver
in the tranched credit market.
51.niapS
Example:
Hedging a credit bond with a CDS.
An investor holds 10 million of Telefonica 5.125 2/2013 bonds with a rating of Baa1/BBB+. This
bond trades at 101.86 giving a yield of 4.75%.
52.niapS
53.niapS
Figure 2-14: Hedging spread (and interest rate) risk of Telefonica 5.125 4/13
In order to hedge 10 million of Telefonica 5.125 2/13 our investor has to buy credit protection
through a CDS. He needs to buy 12.03 million of a 5 year CDS based on current market
information. Should he – in addition to spread risk – also want to eliminate the interest rate risk
then he would additionally enter into an interest rate swap agreement with a nominal amount of
10.64 million (paying fix 4.369 and receiving floating 4.112 6 month Euribor). The screen shot
shows us that a parallel downward movement of the spread curve by 1 bp (DV01) results in zero
movement of the combined bond and CDS position. Likewise, a parallel 1 bp downward
movement of the interest rate curve (IRDV01) will result in an unchanged value of the combined
bond and swap position.
We do not want to enter the technical details of calculation at this point. However it is worth
mentioning that the principle underlying the hedging is based on producing offsetting cash flows
in relation to the instrument to be hedged.
54.niapS
55.niapS
56.niapS
57.niapS
2.6.6 Corporations*
Corporations are relatively new to the credit derivatives markets. They mostly focus on risk
management and increasingly use CDS indices and structured credit products to increase
returns on pension assets or balance sheet cash positions.
58.niapS
The way to value a CDS contract proceeds the same way as for any financial instrument: by
discounting expected future cash flows to the present. The marking-to-market is
approximately equal to the notional amount multiplied by the difference between the contract
spread and the market spread (in basis points per annum) and the risk-adjusted duration of the
contract.
To illustrate this, assume a 5-year CDS contract has a coupon of 150 bps. If the market rallies
to 100 bps, the seller of the original contract will have a significant profit. Assuming a
notional size of USD 10 million the profit is the present value of (150 bps – 100 bps) ∙ USD
10,000,000 or USD 50,000 per year for 5 years. If there were no risk to the cash flows, one
would discount these cash flows by the risk free rate to determine the present value today,
which would be something slightly below USD 250,000. These contracts do have credit risk
however, so the value must be lower than the calculation just described. Assume that the
original seller of the contract at 150 bps chooses to enter into an offsetting contract at 100bp.
This investor now has the original contract on which she is receiving USD 150,000 per year
and another contract on which she is paying USD 100,000 per year. The net cash flow is USD
50,000 per year if there is no default. In the case of default, the contracts cancel each other
leaving the investor with no immediate gain or loss. However she loses the remaining annual
USD 50,000 income stream. The higher the likelihood of a credit event, the more likely it is
that she stops receiving the USD 50,000 payments. Thus the value of the combined short plus
long risk position is reduced.
59.niapS
Standardized confirmation and market conventions greatly improve the marketability. All that
the parties involved need to specify are the terms of the transaction that differ from trade to
trade (e.g., reference entity, maturity date, spread, notional). The ease of transaction is
increased because CDS participants can unwind a trade or enter an equivalent offsetting
contract with a different counterparty from whom they initially traded. As is true with other
derivatives, CDS that are transacted with standard ISDA documentation may be easily
assigned to other parties. In addition, single-name CDS contracts mature on standard quarterly
end dates. These two features have helped promote liquidity and, thereby, stimulate growth in
the CDS market. ISDA’s standard contract has been put to the test and proven effective in the
face of significant credit market stress. When WorldCom filed for bankruptcy in July 2002,
according to estimates, there were 600 CDS contracts outstanding in the marketplace,
accounting for over USD 7 billion in notional terms. More recently, when Parmalat SPA
defaulted in December 2003, it is estimated that there were approximately 4,000 CDS
contracts and EUR 10 billion outstanding in the marketplace. Additionally, Parmalat was a
component of an original credit index. In all situations, the contracts were settled without
mechanical settlement problems, disputes or litigation.
60.niapS
Example:
Investor B buys five years of credit protection (short risk) and pays 50 bp per year. Assume
spreads will widen in one year to 75 bp. Investor B could unwind the CDS by selling 4 year
protection (long risk) at 75 bp. Investor B would receive approximately the present value of the
difference
(75 bp – 50 bp) for the remaining four years multiplied by the notional amount of the swap.
The most liquid maturities are 5 and 10 years and the market is increasingly developing the
liquidity along the whole maturity spectrum. Standard trading sizes are USD 10-20 million
notional for investment grade and USD 2-5 million for high yield credits. In Europe EUR 10
million is typical for high grade bonds and EUR 2-5 million for low grade bonds.
61.niapS
1) Purchase the fixed rate reference bond (asset); this results in a periodic inflow of the
coupon on the bond, which we denote c.
2) Raise floating financing in the amount required to purchase the asset. This results in a
periodic outflow of Libor plus a spread s1, that is, Libor + s1.
3) Enter into an interest rate swap to pay a fixed rate c and receive a floating rate L + s2.
c c long
IRS Investor corporate
Bond
L+s2%
L+s1%
float
financing
This is equivalent to a synthetic CDS position. In the event that the reference instrument does
not default over the life of the contracts, the seller of this synthetic CDS receives a spread s.
In the event of default occurring, the seller of the synthetic CDS undoes all these transactions.
Further, since the reference asset recovers fraction φ of the bond, the seller of the CDS loses
a fraction (1 − φ ) of the fixed rate bond.
Suppose now a market maker sells a CDS (i.e. goes long credit risk) on a given single name.
In order to hedge this position the seller needs to take the opposite position in the synthetic
CDS. That is to say he first shorts the risky bond, deposits the received nominal in a default
free account and contracts a receiver swap (i.e. receive fixed payments and pay floating). This
and the short CDS position will then cancel each other. The market maker will earn the bid-
ask spread.
62.niapS
Viewed in this manner, there is a simple relationship between cash and CDS markets that is
based on the absence of arbitrage. If the CDS spread deviates from s, then, barring transaction
costs and structural differences between markets, one should be able to implement a
convergence strategy that takes advantage of the divergence (i.e. the basis) between the two
markets. Of course, real world complications (which we will not develop further here) make
this more complicated at times.
Similarly to the put-call parity in the options markets one can freely rearrange this equation
with standard algebra to express one instrument as the combination of the others. To obtain a
hedge for the CDS we rearrange this equation to isolate the CDS.
A negative sign means the opposite position in the instrument. We can therefore rewrite this
equation as follows:
Long
Short CDS Defaultable Payer Default Free
= Bond + + Loan
on the Credit Swap
on this credit
A market maker who sold a CDS needs to take the opposite position in comparison to the
right-hand side of this equation: he will short the risky bond and place the proceeds in a
default free deposit and at the same time he will enter into a receiver swap. This and the long
CDS position will cancel each other out and leave the bid-ask spread as the only source of
profit for the dealer.
63.niapS
• In the structural approach, the default is the consequence of some company event such
as its asset value being insufficient to cover a repayment of debt. Such models are
usually extensions of Merton (1974) firm-value model and therefore require
information about the balance sheet of the firm and can be used to establish a link
between pricing in the equity and debt markets.
• In the reduced-form approach, the credit event process is modelled directly by
modelling the probability of the credit event itself – the implied probability of default
is extracted from market prices.
These models are quite complex and we will just illustrate the general principle of one
reduced-form model in a discrete time setup. In the following tree diagram, at each payment
date either the contract has a default event, in which case it ends with a payment of N ∙ (1 −
R) shown in red where R is the recovery rate, or it survives without a default being triggered,
in which case a premium payment of (N ∙ c) / 4 is made, shown in blue. At either side of the
diagram are the cash flows up to that point in time with premium payments in blue and
default payments in red. If the contract is terminated the square is shown with solid shading.
At the ith payment, the probability of surviving over the interval (ti − 1) to ti without a default
payment is pi and the probability of a default being triggered is (1 − pi).
Mid-market CDS spreads can be calculated from default probability estimates (see Hull for
the general principle). For the sake of the exercise, we assume that a reference entity can
default during a year, subject to the condition that it did not default earlier, with a probability
of 3%.
64.niapS
Default Survival
Time (years)
probability probability
1 0.0300 0.9700
2 0.0291 0.9409
3 0.0282 0.9127
4 0.0274 0.8853
5 0.0266 0.8587
Table 2-2: Default probabilities and survival probabilities
Looking from today, the (unconditional) probability of default in year 1 is 0.03. Therefore we
have a probability of 0.97 that the entity will survive until the end of year 1. The probability
of default during the second year is 0.03 ∙ 0.97 = 0.0291 and the probability of survival until
the end of the second year is 0.97 ∙ 0.97 = 0.9409 and so on.
We assume the risk free interest rate (LIBOR) is 3.5% continuously compounded:
exp(-0.035 ∙ 1) = 0.9656, exp(-0.035 ∙ 2) = 0.9324 and so on. ‘s’ is the premium payment per
annum of the CDS. The assumed recovery rate is 20% and we assume that default can only
happen in the middle of the year. Table 2-3 shows the calculation of the expected present
values of the payments made on the CDS at the rate of s per year on a notional of USD 1. For
example there is a 0.8853 probability that the 4th payment of s is made. The expected payment
therefore is (probability ∙ payment) = 0.8853 ∙ s and its present value is 0.8853 ∙
exp(-0.035 ∙ 4) = 0.7696 ∙ s. The sum of all expected payments is 4.1261 ∙ s.
PV of
Expected
Probability Recovery Discount expected
Time (yrs) payoff
of default rate factor payoff
(USD)
(USD)
0.5 0.0300 0.2 0.0240 0.9827 0.0236
1.5 0.0291 0.2 0.0233 0.9489 0.0221
2.5 0.0282 0.2 0.0226 0.9162 0.0207
3.5 0.0274 0.2 0.0219 0.8847 0.0194
4.5 0.0266 0.2 0.0212 0.8543 0.0182
Total 0.1039
Table 2-4: Present value of expected payoffs
As shown in Table 2-4 we assume that a default occurs half-way through the year. There is a
0.0274 probability of default in the fourth year. Given the (expected) recovery rate of 20%,
the expected payoff at this time is 0.0274 ∙ 0.8 ∙ 1 = 0.0219. The present value of the expected
payoff is 0.0219 ∙ exp(-0.035 ∙ 3.5) = 0.0194. The total present value of the expected payoffs
is USD 0.1039.
65.niapS
Expected PV of expected
Probability of
Time (yrs) accrual Discount factor accrual
default
payments payments
0.5 0.0300 0.0150 ∙ s 0.9827 0.0147 ∙ s
1.5 0.0291 0.0146 ∙ s 0.9489 0.0138 ∙ s
2.5 0.0282 0.0141 ∙ s 0.9162 0.0129 ∙ s
3.5 0.0274 0.0137 ∙ s 0.8847 0.0121 ∙ s
4.5 0.0266 0.0133 ∙ s 0.8543 0.0113 ∙ s
Total 0.0649 ∙ s
Table 2-5: Present value of expected accrual payments
In the final step we calculate the accrual payments in case of default. For example there is a
0.0274 probability that there will be a final accrual payment half-way through the third year.
The accrual payment is 0.5 ∙ s. The expected accrual payment at this time is 0.0282 ∙ 0.5 ∙ s =
0.0141 ∙ s. Its present value is 0.0141 ∙ s ∙ exp(-0.035 ∙ 2.5) = 0.0129 ∙ s. The total present
value of expected accrual payments is 0.0649 ∙ s.
The present value of the expected payments is 4.1261 ∙ s + 0.0649 ∙ s = 4.1911 ∙ s. The present
value of the expected payoff is 0.1039. Equating the two we obtain the CDS spread for a new
CDS as: 4.1911 ∙ s = 0.1039 or s=0.0248. Therefore the mid-market spread should be 0.0248
times the principal or 248 basis points per year. At this level of the spread the present value of
the credit default swap to both parties (seller and buyer) is exactly zero, as is usual with most
of the swap contracts.
The above example is simplified. In practice, calculations will be more involved since
payments are made more often than once a year and we might want to assume that defaults
can occur more frequently than once a year.
At its inception, the CDS, like most other swaps, is worth zero. Later its value can change to
become positive or negative. Suppose that the actual CDS spread (on a notional of USD 10
million) shortly after issue declines from the initial 248 bps to 150 bps. This means that the
probability of default has diminished. In fact, using a default probability of 1.827% (while
everything else stays the same) in the Tables 5-3 to 5-5, it can be verified that the present
value of the expected payments becomes 4.312 ∙ s, whereas the present value of the expected
payoff becomes 0.0647. Equating the two we obtain the new CDS spread as: 4.312 ∙ s =
0.0647 or s = 0.0150.
The CDS seller to close out his position could buy a new CDS at 150 bps. He would receive
248 bps and pay 150 bps. The value of swap to the seller is therefore: 4.312 ∙ (0.0248 -
0.0150) ≈ 0.0423 times the nominal (10’000’000 ∙ 0.0423 = 423’000); the value of the swap
to the buyer is -0.0423 times the nominal resulting in -423’000.
66.niapS