Credit Derivatives Overview IIMA

JULY 2006
CREDIT DERIVATIVES
Vaidya Nathan
Outline
Page
Credit Derivatives Overview 1
CLN & Linear Basket 14
Basket Products 26
Synthetic CDO 49
Credit Derivative Options 68
CDS Valuation 87
D E R I V A T I V E S
Credit Derivatives Roadmap 116
Credit Derivatives Update 143

C R E D I T
VAIDYA NATHAN 1
Credit derivatives in the context of financial markets growth
New applications
expanding financial
instruments use
re As
O V E R VI E W
a s
u c t s pa n ex et cl
r od to s ird t
tra ende asses
p
w ging & th cts dit d
e
N er nd du ion bey getti
m o o al o n
e sec pr ma nd g
D E R I VA T I VE S
o n rke
rati ts
ne
ge
C R E D I T
VAIDYA NATHAN 2
Role of Credit Derivatives
Motivations
Motivations for
for use
use of
of Credit
Credit Derivatives
Derivatives
Hedge,
transfer and/or
Synthetically mitigate credit Decompose
create loan- exposure and separate
bond; credit risks
alternative to embedded in
equity financial
derivatives instruments
O V E R VI E W
Generate Proactively
leverage or manage credit
yield risk on a
enhancement Manage portfolio basis
regulatory
D E R I VA T I VE S
capital ratios
C R E D I T
VAIDYA NATHAN 3
Credit derivatives isolate and transfer credit risk
Broad definition
bilateral financial contract which allows specific aspects of credit risk to be

isolated from the other risks of an instrument, and passed from one
counterparty to another
Credit 60 bps
O V E R VI E W
Loan/bond
FX, 6.60 %
Interest yield
D E R I VA T I VE S
Rate
On-balance sheet
C R E D I T
Off-balance sheet
VAIDYA NATHAN 4
Credit derivatives perform a market completion role
Bond = Duration + Convexity + Credit
Convexity Risk
including callability Credit risk
risk sometimes i.e.
negative convexity
O V E R VI E W
D E R I VA T I VE S
Duration Risk
C R E D I T
VAIDYA NATHAN 5
Efficiency gains arising from disaggregating risk
Auctioneer sells a number of

risks, each to the highest bidder
JOB LOT
O V E R VI E W
D E R I VA T I VE S
C R E D I T
VAIDYA NATHAN 6
Spreads of Credit Default Swaps can be compared to bond yields
Bond / Loan Asset Swap Credit Default Swap
Credit Risk
Credit Risk
Funding Risk Credit Risk
Funding Risk
O V E R VI E W
Risk Free Rate

D E R I VA T I VE S
C R E D I T
VAIDYA NATHAN 7
The simplest instrument: single name credit default swaps
Reference Entity
Risk (Notional)
Fee/premium
B A
Protection Buyer Protection Seller
Contingent Payment
upon a credit event
Buy CDS Sell CDS
O V E R VI E W
Buy Protection Sell Protection
“Short Risk” “Long Risk”
Pay periodic payments Receive periodic payments

D E R I VA T I VE S
Receive contingent payment Pay contingent payment

C R E D I T
VAIDYA NATHAN 8
Indicative Summary Terms of CDS
General
General Terms
Terms Fixed
Fixed Payments
Payments
Effective Date: 23 Nov 2006 Fixed Rate Payer Notional: USD 25,000,000
Scheduled Termination Date: 23 Nov 2008 Fixed Rate Payer Payment Dates: The 23rd
of February, May, August and November,
Floating Rate Payer: X (the “Seller”)
commencing on February 23, 2007
Fixed Rate Payer: Y (the “Buyer”)
Fixed Rate: X% per annum
Business Day: London & New York
Fixed Rate Day Count Fraction: Actual/360
Business Day Convention: Modified Following
Reference Entity: ABC

Floating Payment
Floating Payment
O V E R VI E W
Reference Obligation(s) - The obligation(s)

Floating Rate Payer Notional: USD 25,000,000
identified as follows:
Primary Obligor: ABC Corporation Conditions to Payment:
Maturity: 15 September 2011 1. Credit Event Notice
Notifying Party: Buyer or Seller
D E R I VA T I VE S
Coupon: 6.5%
2. Notice of Publicly Available Information
CUSIP/ISIN: USXXX Applicable
Original Issue Amount: USD 1,000,000,000 Public Source(s): Standard Public Sources
Specified Number: Two
3. Notice of Physical Settlement
C R E D I T
VAIDYA NATHAN 9
Indicative Summary Terms of CDS
Credit
Credit Events
Events Settlement
Settlement Terms
Terms
Credit Events: The following Credit Event(s) Settlement Method: Physical Settlement
shall apply to this Transaction:
Physical Settlement Period: Section 8.5 of the
Bankruptcy
ISDA Credit Derivatives Definitions, subject to a
Failure to Pay maximum of 30 Business Days
Restructuring
Portfolio: Exclude Accrued Interest
Grace Period Extension: Not Applicable
Deliverable Obligation Category: Bond or Loan
Payment Requirement: USD 1,000,000 or its
equivalent in the relevant Obligation Deliverable Obligation Characteristics:
Currency Pari Passu Ranking
Specified Currencies: Standard Specified
O V E R VI E W
Default Requirement: USD 10,000,000 or its

Currencies
equivalent in the relevant Obligation
Assignable Loan
Currency
Consent Required Loan
Obligations:
Transferable
D E R I VA T I VE S
Obligation Category: Borrowed Money

Not Contingent
Obligation Characteristics: None
Maximum Maturity: 30 years
Not Bearer
Restructuring Maturity Limitation Applicable
C R E D I T
VAIDYA NATHAN 10
In reality assumes two names risk
Reference Entity Risk
X bps per annum

Counterparty Bank
Contingent Payment
Buyer decreases exposure to Reference Credit(s), but assumes contingent (“two-

O V E R VI E W
name”) exposure to Seller
Seller receives a fee in return for making a Contingent Payment if there is a Credit
Event of the Reference Credit which in turn depends on the financial health of the
D E R I VA T I VE S
bank
C R E D I T
VAIDYA NATHAN 11
Replicating a Credit Default Swap
Corporate Asset
$ 100 T + Corporate Spread (SC)

T + Swap Spread (SS)
Collateral
Investor Swap Market
Libor
Repo rate (L – x)
O V E R VI E W
$ 100 * ( 1 – haircut )
Repo Market
D E R I VA T I VE S
Credit Default Swap Spread (approx.) = Corporate Spread (Sc) – Swap Spread (Ss)
Assumption: Haircut is small ( ≈ 0) & repo rate spread is negligible ( ≈ 0)

C R E D I T
VAIDYA NATHAN 12
Funding Cost Arbitrage
Lender AAA Rated Institution

L – 20 bp
L + 40 bp
BBB Asset
L + 25 bp
Lender AA- Rated Institution
L + 40 bp
BBB Asset
O V E R VI E W
25 bp
AA- Rated Institution AAA Rated Institution
Contingent payment in event of default

D E R I VA T I VE S
L + 40 bp
BBB Asset
Credit Assessment AAA to AA- A+ to A- BBB+ to BBB- BB+ to B- Below B- Unrated

C R E D I T
Risk Weight 20% 50% 100% 100% 150% 100%
VAIDYA NATHAN 13
Outline
Page
Basket Products 26
Synthetic CDO 49
CDS Valuation 87

C R E D I T
VAIDYA NATHAN 14
Degrees of leverage in various Credit Derivative structures
Baskets Portfolio Tranching
Capital protection +
minimum Linear Senior
coupon/interest
Increasing
Yield Capital protection Mezzanine
First-To-Default
Non-capital protected First-To-Default with First Loss (Junior)

Mark-To-Market
B A S K E T
L I N EA R
&
C L N
VAIDYA NATHAN 15
Ab Initio - Credit-linked notes
Tailored notes
Structured for investor’s required currency, maturity, and coupon needs
CLNs can be rated and / or listed if required
Both physical and cash delivery are available to the investor
Provide solutions for many investors restricted from entering into OTC transactions
Provide investors with yield and minimum ratings requirements through leveraged
high grade structures
B A S K E T
L I N EA R
&
C L N
VAIDYA NATHAN 16
Structure of a Typical CLN
Interest Rate Swap
Protection Sale
Proceeds
Swap Counterparty SPV Collateral
CDS Premium
CLN Proceeds
Investors
B A S K E T
L I N EA R
&
C L N
VAIDYA NATHAN 17
Linear Baskets
Linear basket swaps allow investors to gain exposure to multiple credits in one
trade
Risk buyer takes on exposure to each credit equal to the 1/N of the notional of the
basket, where N is the number of credits in the basket (assuming equal weighting)
After the first credit event:

swap on the defaulted credit terminates,
notional of the trade is reduced by the notional of the defaulted credit,
the investor bears exposure to the non-defaulted credits
Yield on these structures is additive, since each credit is independent of the other
(the same as yield on first-to-default basket with zero correlation)
Advantage of less documentation by taking exposure to many credits in one single

B A S K E T
trade
L I N EA R
&
C L N
VAIDYA NATHAN 18
Advantages of Credit Linked Notes
CP Risk
CP Risk &
& No
No Direct
Direct
Credit Line
Credit Line Derivatives
Derivatives
Usage
Usage Contract
Contract
No
Nosystem
system Non-issuers
Non-issuers
requisites
requisites reference
reference
CLN Advantages
Canbe
Can be Customized
Customized
listed
listed Maturity
Maturity
Relative
Relative Tailored
Tailored
B A S K E T
Value
Value Exposure
Exposure
L I N EA R
&
C L N
VAIDYA NATHAN 19
Disadvantages of CLNs
Lack of Cheapest to
Liquidity Deliver
Option
Medium
Funded Form
Term
& Lack of
View
Leverage
B A S K E T
L I N EA R
&
C L N
VAIDYA NATHAN 20
Linear Baskets - an illustration
X bp per annum on notional 100
A B C D E Investor
Investor
Contingent Payment
(Par — Recovery on Credit E)
Example: Green Bottle Swap on 5 equally weighted names on a notional of 100

If Credit E defaults, Protection Seller pays Par and receives the Recovery Value on Credit E
The rest of the swap remains, with the notional falling to 80
X bp per annum on notional 80

B A S K E T
A B C D Investor
Investor
Contingent Payment
L I N EA R
(Par — Recovery on defaulted credit) Protection Seller

Protection Buyer
&
C L N
VAIDYA NATHAN 21
Linear Basket - Example
300 bps A
100 bps
B Equal weighted Linear Basket Spread
= (300 + 100 +400 + 150 +200)/5
=230 bps
400 bps
C
B A S K E T
150 bps
D
L I N EA R
200 bps E
&
C L N
VAIDYA NATHAN 22
Benefits of CDX
Use Credit Default Swaps to maximize liquidity -

Liquidity
Liquidity
portfolios are composed of the most liquid credit
default swap names
Cost efficient and timely access to the Credit Markets

Diversification
Diversification via index swaps and credit-linked securities
Daily reports on actual versus theoretical pricing

Transparency
Transparency
B A S K E T
L I N EA R
&
C L N
VAIDYA NATHAN 23
CDS indices: CDX and iTraxx
Two major CDS indices trade actively at tight bid-ask spreads
DJ CDX in North America has 125 names
DJ iTraxx Europe has 125 names
Weekly fixings on three CDS indices: DJ iTraxx Europe, HiVol index and
Crossover index
Can also trade loss tranches on index
Tranches are like synthetic CDO tranches
Just as options are way to trade volatility, tranches are way to trade
default correlations
B A S K E T
One-factor Gaussian copula is standard for quoting correlations

L I N EA R
&
C L N
VAIDYA NATHAN 24
CDS indices for Asia and Australia
Nonlinearity
Nonlinearity of
of risky
risky duration
duration for
for half
half and
and double
double credit
credit spreads
spreads
iTraxx CJ has 50 Japanese names, with sub-indices for capital goods, tech
and HiVol
iTraxx Asia has 30 names from outside Japan with sub-indices for Korea,
Greater China and rest of Asia
iTraxx Australia has 25 names
CDX.EM has 14 emerging market sovereigns, including Korea, Malaysia and

the Philippines
B A S K E T
L I N EA R
&
C L N
VAIDYA NATHAN 25
Outline
Page
Basket Products 26
Synthetic CDO 49
CDS Valuation 87

C R E D I T
VAIDYA NATHAN 26
First To Default (FTD) Baskets
First-to-default basket swaps allow investors to leverage their exposure to a

basket of credits
After the first credit event, the first-to-default swap terminates and the investor no
longer bears exposure to the non-defaulted credits
Yield enhancement in these structures basically depends on the correlation of the

names in the basket
Risk buyer takes on exposure to each credit equal to the notional of the basket,
thus achieving leverage of the number of names in the basket
P R O D U C T S
B A S K E T
VAIDYA NATHAN 27
First To Default (FTD) Baskets — an illustration
X bp per annum
A B C D E
Contingent Payment
Protection Buyer (Par - Recovery on Credit E) Protection Seller
If Credit E defaults, Protection Seller pays Par and receives the Recovery Value on
Credit E
The first-to-default swap is terminated and Protection Seller has no further

exposures
The greater the correlation, the greater the probability of multiple defaults in the
P R O D U C T S
basket
B A S K E T
VAIDYA NATHAN 28
Mechanics of a First To Default Basket Structure
Risk is sourced from the market
First to Default Tranche is sold either in bond or swap form
Bank retains “Senior” Tranche
Bank
Market Credit Default
Swaps on
5 CDS on 1st to
Individual Default
Credits
OTC
Investors
P R O D U C T S
B A S K E T
VAIDYA NATHAN 29
From individual default probability to basket default probability
Individual
Individual Basket default probability
default
default Correlation = 0
probability
probability 150 bps A FTD Spread = 575 bps
Correlation
Correlation Taking a leveraged exposure to
of
of a basket is equivalent to
assets
assets trading the correlation
between those names
50 bps
B
Intuition:
Intuition: The
The higher
higher the
the correlation,
correlation, the
the lower
lower the
the
spread
spread on
on the
the leveraged
leveraged piece
piece
100% correlation
200 bps
C
The basket behaves like 1 single credit
Protection Seller will expect to receive the
widest of individual spreads Correlation = 1.0
FTD Spd = 200 bps

0% correlation 75 bps
D
P R O D U C T S
Each name in the basket behaves

independently. Protection Seller should
receive the sum of the individual spreads
100 bps E
B A S K E T
VAIDYA NATHAN 30
High and Low Correlation – the Tom & Jerry way
High
High &
& Low
Low Correlation
Correlation
P R O D U C T S
B A S K E T
VAIDYA NATHAN 31
Risk illustrated
Correlation = 0
A B
D
C E
Correlation = 1
B
E
C
A
0 < Correlation < 1

P R O D U C T S
A
B E
C D
B A S K E T
VAIDYA NATHAN 32
Actual FTD trades – Example 1
Background
Background Trade Summary
Trade Summary
Korean client buys FTD loan on six Aggregate bid spread: 537 bps
investment grade credits
FTD Basket coupon: 6.75%
50% of names chosen were local credits
FTD spread over Libor: 335 bps
50% of names were foreign credits
FTD spread over Libor as a % of
Inclusion of foreign credits helps aggregate spread: 62%
reduce correlation of the basket which
FTD spread over Libor as a % of highest
inturn helps increase spread
bid spread: 163%
Non callable
Credit 5-year (bids)

Kookmin Bank 62
KEPCO 54
P R O D U C T S
POSCO 53
Hutchison Whampoa 98
Ford 206
Standard Life Assurance 74
B A S K E T
VAIDYA NATHAN 33
Background
Trade Summary
Client buys FTD note on five high yield Aggregate bid spread: 950 bps
credits
FTD Basket coupon: 10.60%
Names chosen were high rated (BB)
FTD spread over Libor: 720 bps
high yield credits
Clustered spreads
aggregate spread: 75%
Spreads have low correlation to
maximize spread
bid spread: 335%
Non callable
Credit 5-year (bids) Rating

Amerisource Corporation 160 Ba3/BB
P R O D U C T S
Chesapeake Energy Corporation 215 Ba3/BB-

Mandalay Resort Group 210 Ba2/BB+
Flextronics International Ltd 167 Ba2/BB-
Georgia Pacific Corporation 208 Ba2/BB+
B A S K E T
VAIDYA NATHAN 34
Background
Trade Summary
Client buys FTD protection on five high Aggregate offer spread: 795 bps
yield credits
FTD Basket premium: 5.0%
Reduced costs relative to hedging each
of the individual credits
aggregate spread: 63%
Sheds substantial portion of the risk
Credits have high correlation to bid spread: 222%
minimize cost
Credit 5-year (bids)

Delphi Corporation 113
P R O D U C T S
Ford Motor Company 213

General Motors Acceptance Corporation 157
Lear Corporation 87
Visteorn Corporation 225
B A S K E T
VAIDYA NATHAN 35
Default Correlation
Default correlations are key determinants of hedge ratios which determine basket
premiums that dealers are willing to pay. The boundary conditions for the basket
premium can be restated in terms of the default correlation as follows:
Basket premiums should decline with an increase in correlation. A basket of

uncorrelated credits trading at similar spreads produces the largest relative
increase in premium compared to the average single-name default swap premium
Default correlations impact the likelihood of multiple defaults up to a given time

horizon. In practice, there is a lack of historical data that could be used to extract
default correlations. Instead, market players use the asset correlation to calculate
default correlation
Asset correlations can be extracted from the ability-to-pay process of a portfolio of

firms. Such a process is modeled for an individual firm as its market value of assets
minus liabilities. Market inputs are equity and debt data. The asset correlation
derived in this manner is deterministically related to the default correlation, i.e.
one can be transformed into the other
P R O D U C T S
The most difficult factor to incorporate in the model for pricing basket products is
the estimation of the underlying correlations. Estimating the correlation between
two default events cannot be achieved by standard statistical methods. Instead
B A S K E T
some proxy for default or credit behaviour must be used

VAIDYA NATHAN 36
How to deal with correlation?
Model the correlation between names as a result of the correlation of each name with
a systemic “market” variable (CAPM)
given a market environment, the names default independently
the probability of default of any given name depends on the market environment
As correlation approaches zero: FTD Basket Premium = Sum of Basket Component’s

CDS Spreads
As correlation approaches one: FTD Basket Premium = Highest Individual CDS Spread
in Basket
Investors wishing to maximize premium generated by selling FTD Basket protection

should consider selecting a basket that is relatively uncorrelated
P R O D U C T S
B A S K E T
VAIDYA NATHAN 37
Par spreads for FTD with different betas
Par
Par spreads
spreads for
for FTD
FTD with
with different
different betas
betas
550
Par Spreads with Time for various betas
500
450
400
350 Rate of decrease of spreads is high for high correlation
Beta = 10% Beta = 50% Beta = 90% Max Spread

300
250
200
Time ( yrs)
150
P R O D U C T S
1.00 2.00 3.00 4.01 5.01 6.01 7.01 8.01 9.01 10.01 12.01 15.01 20.02 25.02 30.02 40.03
Basket of five names each trading at 100 bps. Rate of decrease of spreads is high for high correlation
B A S K E T
VAIDYA NATHAN 38
Expected loss for FTD with different betas
Expected
Expected loss
loss for
for FTD
FTD with
with different
different betas
betas
100.00%
Expected Loss for FTD with time for various betas
90.00%
80.00%
Beta = 10% Beta = 50% Beta = 90%
70.00%
60.00%
50.00%
40.00%
30.00%
20.00%
10.00%
Time (yrs)
0.00%
P R O D U C T S
0.25 1.00 1.50 2.00 3.00 4.01 5.01 6.01 7.01 8.01 9.01 10.01 12.01 15.01 20.02 25.02 30.02 40.03
B A S K E T
VAIDYA NATHAN 39
Dynamic Hedging Of The FTD Basket
The hedging behaviour of a dealer provides some intuition behind the actual basket
premium
A dealer that buys protection on a basket from an investor would normally hedge
this transaction by selling default protection on each individual name in the basket
As the underlying default premiums shift, the deltas will change and the hedges
will need to be rebalanced dynamically
The efficiency with which the hedge can be managed is a key factor that
determines the basket premium
For small movements in the hedge ratio, the dealer may not be able to sell or buy
protection and may instead buy or sell bonds to hedge, thus taking on basis risk
P R O D U C T S
B A S K E T
VAIDYA NATHAN 40
Dynamic Hedging Of The FTD Basket
X bp per annum
A B C D E FTD
FTDInvestor
Investor
Contingent Payment
(Par - Recovery on Credit E)
Notional = N
AANotional
NotionalNNAA==Hedge
HedgeRatio
RatioAAxxNN
BBNotional
NotionalNNBB==Hedge
HedgeRatio
RatioBBxxNN
CCNotional
NotionalNNCC==Hedge
HedgeRatio
RatioCCxxNN
P R O D U C T S
DDNotional
NotionalNNDD==Hedge
HedgeRatio
RatioDDxxNN
EENotional
NotionalNNEE==Hedge
HedgeRatio
RatioEExxNN
B A S K E T
VAIDYA NATHAN 41
Dynamic Hedging of FTD Basket
Following a credit event, the dealer will be forced to unwind

the hedges on the other credits (assuming non-zero deltas for
these credits) 150 bps A
The cost of unwinding the hedge would depend on the spread
movement for each of the non-defaulted credits. This, in turn,
would depend on the correlation between the defaulted and
50 bps
B
the non-defaulted credits
The greater this correlation, the greater the expected spread

widening for a non-defaulted single-name default swap. This
would imply a greater cost of unwinding the hedge. The dealer
200 bps
C
would therefore maintain a lower delta i.e. sell a lower amount
of protection, to minimise losses from the unwind. This would,
in turn, provide a lower premium to pay for the basket
protection
75 bps
D
P R O D U C T S
On the other hand, a low correlation would imply a lower

expected spread change in a non-defaulted credit in the event
of default and consequently a lower cost of unwinding that
hedge. The hedger could therefore maintain a higher delta to
100 bps E
B A S K E T
manage the hedge i.e., sell a higher amount of protection. This

provides a higher premium to pay for the basket protection
VAIDYA NATHAN 42
Investment rationale for a FTD Basket Swap
Efficient leverage and limited downside : Investors are able to efficiently leverage
credit risk with a defined downside potential by executing a FTD Basket Swap, since
the swap’s notional amount references a basket that is 5x to 10x its size
Limited maximum loss: a $10mm FTD Basket Swap that references a basket of ten
names (aggregate credit exposure of $100mm) still limits the investor’s maximum
loss to the first loss in the basket — or $10mm
Basket component correlation premium : FTD baskets that are relatively

uncorrelated pay investors a premium that approaches the sum of the basket’s
component default premiums rather the spread of the basket’s riskiest component
Efficient usage of regulatory capital: The notional amount of the swap, rather than
the notional amount of the reference basket, would be used in a risk-based capital
requirement calculation for a US bank. This allows banks to take credit risk to a
reference portfolio that is 5x-10x as large as the notional amount of the swap for
P R O D U C T S
which it must set aside capital

B A S K E T
VAIDYA NATHAN 43
Investment rationale for a FTD Basket Swap
Investment grade ratings available: Rating agencies have been willing to provide
investment grade ratings to FTD Basket Swaps that meet rating agency requirements
regarding credit risk and diversification
Fund managers with limits on scope of assets to invest: This has allowed investors
that are limited to specific ratings-based investment guidelines to earn premiums
well in excess of those typically found in the single-name investment grade bonds or
credit default swaps
Executable via swap or funded note/deposit: Investors may elect sell FTD Basket
protection via a swap or funded credit linked note (CLN)
The settlement mechanics of a FTD Basket Swap are identical to a single name
credit default swap (physical settlement of defaulted securities)
P R O D U C T S
B A S K E T
VAIDYA NATHAN 44
First 2 To Default (F2TD) Baskets
First-2-to-default basket swaps allow investors to leverage their exposure to a

basket of credits to a lesser extent
The protection buyer is protected against the first two defaults
Total premium for a F2TD basket should be lesser than for a first to default
After the first credit event, contract settles partially and notional of the trade is
reduced by half
After the second credit event, contract settles fully and the trade terminates
P R O D U C T S
B A S K E T
VAIDYA NATHAN 45
First 2 To Default (F2TD) Baskets
X bp per annum
A B C D E
Contingent Payment
Protection Buyer (Par - Recovery on Credit E) Protection Seller
If Credit E defaults, Protection Seller pays Par and receives the Recovery Value on Credit E
X bp per annum
A B C D
Contingent Payment
Protection Buyer (Par - Recovery on Credit D) Protection Seller
P R O D U C T S
If Credit D defaults, Protection Seller pays Par and receives the Recovery Value on Credit D
Trade terminates
B A S K E T
VAIDYA NATHAN 46
First-to-default vs First-2-to-default
Basket Spreads
200
180 180
170
160 158
BPS
145
140 140
135
125
120
112
100
92
80
4Y 5Y 8Y
F2TD FTD1 FTD2

P R O D U C T S
FTD1 : Pacific Dunlop, Pasminco, MIM, Mayne Nickless, Qantas

FTD2 : CWOptus, Boral, United Energy, Woodside, Western Mining
F2TD : FTD1 + FTD2
B A S K E T
VAIDYA NATHAN 47
Other Basket default products
Second to default : The protection buyer is protected only against the second default,
and pays a premium until the occurrence of this event. The premium will be less than for
a FTD, since two defaults are always less likely than one (assuming that the credits are
not perfectly correlated)
Second, third, fourth or fifth to default : The premium for an nth to default decreases
as n increases, since the probability of n defaults becomes more unlikely. The effect of
correlation is to increase the premium, since this increases the probability of multiple
defaults. Relatively speaking, this effect is more dramatic for larger n
Sum of first, second, third, fourth and fifth to default : The total premium for these
five contracts is always greater than the sum of the individual default swap premiums
Quanto basket default swap : Similar to other quanto products. Assume that the FTD
basket is cash-settled. Upon a credit event, observe the reference obligation (presumably
denominated in USD), then protection buyer receives 1-R x Quanto FX i.e. an FX rate
agreed
P R O D U C T S
Actual Quanto Trade: a Taiwanese investor buys protection and pays a running TWD
fee. The basket consists of 5 names with reference obligations denominated in USD.
There is a credit event, recovery on the deliverable is 40. The Taiwanese investor gets
B A S K E T
(100-40) x whatever FX rate that was agreed at the beginning of the trade. For trades in
general, the fee and the payout don't have to be the same currency
VAIDYA NATHAN 48
Outline
Page
Basket Products 26
Synthetic CDO 49
CDS Valuation 87

C R E D I T
VAIDYA NATHAN 49
A synthetic tranche is a specific allocation of risk of a synthetic CDO
A synthetic CDO pools the risk of various synthetic assets — corporates, sovereigns,
financials — in the form of a portfolio of CDS
Credit risk is re-allocated from individual credits to tranches
Equity owner is in the first loss position
Subordinated tranches (including Equity) are leveraged exposures
Senior tranches (AAA, AAAA) are de-leveraged exposures
Each tranche exhibits different risk characteristics which reflect its expected share of
portfolio losses as well as its sensitivity to changes in the expected distribution of
losses within the portfolio
An active risk management strategy allows banks to create and offer tranche
protection or exposure on a standalone basis
C D O
S Y N TH E T I C
VAIDYA NATHAN 50
9
Each tranche is defined in terms of pay-off
Example
Example of
of a
a mezzanine
mezzanine tranche
tranche
Tranche
loss
CDS Buyer of CDS Seller of
Protection Protection
Fee
Tranche
loss
Tranche size
Portfolio loss
Tranche size
C D O
S Y N TH E T I C
VAIDYA NATHAN 10
51
Sum of risks (losses) is equal to portfolio risk (loss)
Tranche
loss
Senior
Portfolio loss
Eq. size Mezz. size
Tranche
loss
Mezz. size Mezzanine
Portfolio loss
Mezz. size
Tranche
loss
Eq. size Equity
Portfolio loss
Eq. size
C D O
S Y N TH E T I C
VAIDYA NATHAN 52
11
Risk of a tranche is defined by its position in the capital structure
Portfolio Tranched structure
Individual Third-loss piece De-leveraged

credit default (Senior piece)
swaps
Second-loss piece
Medium risk
(Mezzanine)
First-loss piece Highly leveraged

(Equity)
C D O
S Y N TH E T I C
VAIDYA NATHAN 53
12
Synthetic first loss tranche is created by sourcing delta amount of CDS
As in a CDO, risk is sourced from the CDS market; however, the amount of risk
sourced reflects only the risk of the first loss tranche
By synthetically reproducing its specific risk profile, any tranche can be created
without the need to place the remainder of the capital structure
Risk This risk is managed through active

rebalancing of the 1st Loss delta
Market Bank
Spread
Risk on 1st Loss Risk on 1st Loss

Tranche Tranche
Delta amount
of CDS on SPV Investor
individual
Spread Spread
credits
Delta is specific to the risk of the

1st Loss Tranche Collateral
C D O
S Y N TH E T I C
VAIDYA NATHAN 54
13
Pricing framework for a synthetic tranche is based on expected loss
At the portfolio level, spread compensates for expected loss:
Portfolio Expected Loss = PV (Portfolio Credit Spread)
Risk Compensation for Risk
Tranching reallocates losses within the portfolio
The spread on an individual tranche must compensate for the tranche

expected loss
Tranche Expected Loss = PV (Tranche Credit Spread)

C D O
S Y N TH E T I C
VAIDYA NATHAN 55
15
To compute expected loss, obtain portfolio loss distribution
Ingredients for the calculation of the loss distribution:

Recovery rate for each name — use CDS market estimates
Probability of default for each individual name — derived from the credit spread
and recovery rate estimate:
PD = S/(1-Recovery Rate)
Default correlation between names — derived from historical asset correlations

C D O
S Y N TH E T I C
VAIDYA NATHAN 56
16
Estimating loss distribution requires tranching, spreads & recoveries
A
A simplified
simplified example
example of
of a
a portfolio
portfolio containing
containing only
only one
one name
name
Tranched structure
Portfolio:
Spread (bps) Notional

150 100 60%— Senior
Maturity = 1 year 30%— Mezz
Recovery Rate Assumption = 50% 10%— Eq

C D O
S Y N TH E T I C
VAIDYA NATHAN 57
17
Probability-weighting loss scenarios gives expected loss of tranches
Portfolio
Portfolio loss
loss distribution
distribution
Eq Mezz Senior
0% 50% 100%
Tranche pricing
Tranche pricing
Not ional Port folio loss Mezz Sr
Scenarios Probabilit y default e d Port folio loss (%) Equit y loss loss loss
1 97. 00% 0 0 0% 0% 0% 0%
2 3. 00% 100 50 50% 10% 30% 10%
Whole port folio Equit y Mezz Senior

Expe ct ed loss (% of Not ) 1. 5% 0. 3% 0. 9% 0. 3%
Tranche Not ional 100% 10% 30% 60%
Tranche Spread (bps) 150 300 300 50
C D O
S Y N TH E T I C
VAIDYA NATHAN 58
18
To model loss distribution, we take into account default correlation
Credit Spreads Recovery Rates
Asset Market Implied

Correlations Default Probabilities
Default Correlations
Model
Loss distribution
C D O
S Y N TH E T I C
VAIDYA NATHAN 59
19
Address default correlation in estimating the loss distribution?
We model the correlation between names as a result of the correlation of each

name with a systemic market variable (similar to CAPM)
Given a market environment, the names default independently
However, the probability of default of any given name depends on the market
environment
Integrating across all possible market environments yields a loss distribution which
then incorporates the correlation between names
C D O
S Y N TH E T I C
VAIDYA NATHAN 60
20
Probability of default depends on the market environment
1-year default
1-year default probability
probability Specific
Specific market
market scenarios
scenarios
Market Bad Neutral Good
environment
Probability
1/3 1/3 1/3
of market
1 year def. Prob 1 year def. prob 1 year def prob
for “Bad” for “Neutral” for “Good”
(Average over markets)
Name 1 1.0% Name 1 1.90% 1.00% 0.10%
Name 2 2.0% Name 2 3.80% 2.00% 0.20%
Name 3 0.5% Name 3 0.50% 0.50% 0.50%
Name 3 is uncorrelated with the market

C D O
S Y N TH E T I C
VAIDYA NATHAN 61
21
Final loss distribution is calculated by averaging over all scenarios
Probability
Probability of
of loss
loss
Bad
market times 1/3
Loss
Neutral
market times 1/3
Good times 1/3

market Loss
C D O
S Y N TH E T I C
VAIDYA NATHAN 62
22
Loss distribution changes with spreads, recoveries, and correlation
Probability
Probability of
of loss
loss
Senior
when correlation increases
when correlation decreases
when spread decreases when spread increases Loss
Average loss
C D O
S Y N TH E T I C
VAIDYA NATHAN 63
23
Portfolio loss distributions for two portfolios
Tranche
Tranche loss
loss different
different for
for same
same tranche
tranche size
size for
for two
two different
different portfolios
portfolios
C D O
S Y N TH E T I C
VAIDYA NATHAN 64
23
As loss distribution changes, delta of any specific tranche changes
In a fully-sold synthetic CDO, the risk of changes in the loss distribution is passed
completely to investors
The distribution of losses within the portfolio will fluctuate, but the portfolio
hedge is static
In a synthetic tranche, bank’s hedge must offset the risk position of the specific
tranche
Initial delta reflects the relative proportion of the portfolio loss distribution
which falls within the tranche
Therefore, as the tranche expected loss changes, the delta hedge for that tranche
must also be rebalanced
C D O
S Y N TH E T I C
VAIDYA NATHAN 65
24
Risk management for synthetic tranches is rapidly evolving
New technology allows for the creation of synthetic tranches with a range
of attractive characteristics beyond the original static
Tranche exposure can be offered in local currency whereas risk is sourced

in a foreign currency
Term of tranche credit exposure can differ from the term of the note or
swap
Credit exposure of coupons can differ from the exposure of the principal
Investor or asset manager can be offered limited portfolio management

flexibility
C D O
S Y N TH E T I C
VAIDYA NATHAN 66
26
Credit derivatives offer some unique characteristics
Credit derivatives:
unbundle credit risk from other aspects of ownership (tax,

accounting, liquidity, relationship)
are similar in substance to many traditional credit instruments
are not triggered by underlying price movements but only by default
provide the only efficient short positioning vehicle
frequently require explicit consideration of correlation risk
are available on- or off-balance sheet and can provide leverage

efficiently
C D O
S Y N TH E T I C
VAIDYA NATHAN 67
Outline
Page
Basket Products 26
Synthetic CDO 49
CDS Valuation 87

C R E D I T
VAIDYA NATHAN 68
Options on Credit Default Swaps
Options on credit default swaps can be structured into trades that work for various
parties with different objectives
Fund managers can sell options to earn premium while waiting for spreads to
widen to their target investment levels
Hedge funds can sell volatility through options to earn premium
Investors can buy callable CLNs/CDS to earn higher returns than on plain-vanilla
CLNs/CDS
Investors can buy principal protected notes that are linked to spread tightening
or spread widening options or their combinations
Investors can combine CDS options with interest rate or equity products to earn
OP TI ON S
higher returns or reduce their risk profile on their investment

DE R I VAT I VE
CR E D IT
VAIDYA NATHAN 69
Options terminology
A call on a credit default swap is the right to buy risk/sell protection (receiver
option)
A put on a credit default swap is the right to sell risk/buy protection (payer option)
A straddle is a combination of a put and a call at the same strike
ATM denotes an option struck at-the-money
OTM denotes an option struck out-of-the-money

OP TI ON S
DE R I VAT I VE
CR E D IT
VAIDYA NATHAN 70
Options payout diagrams
Bullish Strategy Bullish Strategy
Decreasing Price Increasing Price Decreasing Price Increasing Price

Increasing Spreads Decreasing Spreads Increasing Spreads Decreasing Spreads
80 bps 50 bps 20 bps 80 bps 50 bps 20 bps

Long call
Short put
(unlimited upside when spreads tighten (unlimited downside when spreads widen,
downside limited to premium paid) upside limited to premium earned)
Short call
Bearish Strategy Bearish Strategy
OP TI ON S
(unlimited downside when

Long put
spreads tighten, upside limited to
(unlimited upside when spreads widen
premium earned)
downside limited to premium paid)
DE R I VAT I VE
Decreasing Price Increasing Price Decreasing Price Increasing Price

Increasing Spreads Decreasing Spreads Increasing Spreads Decreasing Spreads
80 bps 50 bps 20 bps 80 bps 50 bps 20 bps

CR E D IT
VAIDYA NATHAN 71
Options on single name credit default swaps
Underlying: 5yr CDS on a single Reference Entity
Maturity: 3 months
Premium: prices are quoted in cents, premium is paid T+3 business days
Strike: At-the-money-spot spread of the 5yr CDS derived from the current trading
level at the time of the trade
Exercise: European (only at the maturity of the option)
Settlement:
Physical: into the 5yr CDS upon exercise at the pre set strike
OP TI ON S
Cash: option is unwound upon exercise and cash paid out to client
Knock out:
DE R I VAT I VE
Both calls and puts knock out in case of a Credit Event on the Reference Entity
CR E D IT
VAIDYA NATHAN 72
Options on TRAC-X Europe
TRAC-X Europe options are a liquid, standard product to trade credit volatility on a
macro basis
Underlying: TRAC-X Europe 99 Swaps with 20 September 2011 maturity
Maturity: 3 months or 6 months
Premium: prices are quoted in cents, premium is paid T+3 business days
Strike: Preset strikes of 30, 35, 40, 45, 50 and 55, chosen to reflect ATM and OTM
forward levels in increments of 5 bps
Exercise: European (only at the maturity of the option)

OP TI ON S
Settlement: Physical, into TRAC-X Europe 99 Swaps upon exercise at the pre set
strike
DE R I VAT I VE
Knock out: No knock out in case of a Credit Event in TRAC-X Europe 99 Swaps
CR E D IT
VAIDYA NATHAN 73
CR E D IT DE R I VAT I VE OP TI ON S
TRAC-X
VAIDYA NATHAN
74
Trade Idea 1: Long credit trade to earn premium
A fund manager who finds France Telecom too tight compared to his target investment levels
could sell an OTM put on France Telecom to earn premium
Manager receives 15 cents

15 ¢
to sell OTM put
Put ATM:
strike: 66 bps
Decreasing Price Increasing Price
Increasing Spread 76 bps Decreasing Spread
BE:
80 bps
OP TI ON S
Breakeven analysis:
DE R I VAT I VE
Investor receives 15 cents to sell a 3-month European put on France Telecom struck at 76 bps
If France Telecom widens by more than 4 bps (= 15 bps / 4.3 duration) from the strike in 3
months, the investor will lose money on the option (I.e., if France Telecom widens beyond 80
bps, the investor will lose more on the option than the option premium earned)
— However, the manager will be put into France Telecom risk at a strike corresponding to
CR E D IT
his target investment level
VAIDYA NATHAN 75
Trade Idea 2A: Short credit trade
An investor who believes the market will widen in the short to medium term could buy an ATM
put on TRAC-X Europe to benefit from spreads widening
Put strike:
ATM: 45 bps Increasing Price
Decreasing Price
Increasing Spread Decreasing Spread
BE:
53 bps
36 ¢ Investor pays 30 cents
to buy ATM put
Breakeven analysis:
OP TI ON S
Investor pays 36¢ to buy a 3-month maturity European put on TRAC-X Europe struck at 45 bps
If TRAC-X Europe widens by more than 8 bps (= 36¢ / 4.3 duration) from the strike in 3 months, the
investor will make money from exercising his ATM put (i.e., if TRAC-X Europe widens beyond 53
DE R I VAT I VE
bps, the investor will make more on the option than the option premium paid)
If TRAC-X Europe widens to 60 bps in 3 months, the investor will make 64.5¢ (= (60-45) bps x 4.3
duration) from exercising his ATM put. The investor’s payout ratio in this case will be 1.79 (= 64.5¢
payout / 36¢ option premium paid)
CR E D IT
If TRAC-X Europe widens to 70 bps in 3 months, the investor will make 1.1% (= (70-45) bps x 4.3
duration) from exercising his ATM put. The investor’s payout ratio in this case would be 2.99 (=
1.1% payout / 36¢ option premium paid) VAIDYA NATHAN 76
Trade Idea 2B: Short credit trade with capped upside
If the same investor finds the ATM put too expensive and thinks market spreads will widen but
not by too much, he could buy an ATM put and sell an OTM put to subsidise cost of the ATM put
Bear spread payout

Put Put
strike: strike:
60 bps ATM: 45
Increasing Spread BE: Decreasing Spread
bps
45 bps
30 ¢ Investor pays 21 cents
to buy bear spread
Breakeven analysis:
OP TI ON S
Investor pays 30¢ net to buy a 3-month maturity European put on TRAC-X Europe struck at
45 bps and sell a 3-month European put on TRAC-X Europe struck at 60 bps
If TRAC-X Europe widens by more than 7 bps (= 30¢ / 4.3 duration) from 50 bps in 3 months,
DE R I VAT I VE
the investor will make money on the ATM put he bought (i.e., the upside from buying the
option will be higher than the option premium paid)
Maximum payout on this trade will be 64.5¢, if TRAC-X Europe goes to 60 bps or wider (=
{60 bps – 45 bps} x 4.3 duration). The investor’s payout ratio in this trade is 2.15 (= 64.5¢ /
30¢ option premium paid)
CR E D IT
If investor thinks the market will not widen beyond 60 bps, he should put on the bear
spread (payout ratio of 2.15) instead of buying the ATM put outright (payout ratio of 1.79
when spreads are at 60 bps) VAIDYA NATHAN 77
Trade Idea 3: Short volatility trade
A hedge fund does not have a strong credit view but believes that credit market will
experience little volatility in the short to medium term could trade volatility and
earn premium by selling a 3-month ATM straddle on TRAC-X Europe
36 ¢
Investor receives 36 cents
to sell straddle
B/E on put: ATM: B/E on call:

53 bps 45 bps 37 bps
Increasing Spread Decreasing Spread
OP TI ON S
Breakeven analysis:
Hedge fund receives 36 cents to sell a 3-month maturity European straddle on
DE R I VAT I VE
TRAC-X Europe struck at 45 bps

If TRAC-X Europe tightens or widens by more than 8 bps (= 36¢ / 4.3 duration)
from the strike in 3 months, the fund will lose money on the option (i.e., the
downside from selling the option will be higher than the option premium earned)
CR E D IT
VAIDYA NATHAN 78
Trade Idea 4: Callable CDS
A CDS investor with investment targets that are higher than current market levels could sell 10yr France
Telecom protection callable by bank in 5 years that pays more than the 10yr plain-vanilla CDS
Callable CDS
Premium for 10yr FRTEL CDS callable in 5yrs
If 5yr FRTEL CDS < 95 bps in 5yrs

Bank Bank calls the CDS from the investor Investor
If 5yr FRTEL CDS > 95 bps in 5yrs
Investor holds the CDS contract for 10yrs

OP TI ON S
Breakeven analysis:
5yr France Telecom CDS: 66 bps, 10yr France Telecom CDS: 89 bps
DE R I VAT I VE
10yr France Telecom CDS callable in 5 years pays 95 bps (WHAT IS THE CATCH HERE?)
If Bank calls the CDS in 5 years, investor will have earned 29 bps (44%) more running than the
5yr CDS, and FRTEL 5yr CDS would have to tighten by more than 29 bps for investor to lose
the 29 bps earned in the first 5 years to enter into a new France Telecom CDS
CR E D IT
If Bank does not call the CDS, investor will have earned 6 bps (7%) more running than selling
plain-vanilla 10yr protection
VAIDYA NATHAN 79
Binary (Digital) Credit Swaps
Z bps per annum
Contingent Payment
(Par)
Clients looking for leverage opportunities on a single name can sell protection in a binary swap
In a plain-vanilla credit swap, the protection seller would receive the recovery value of the
OP TI ON S
Reference Credit in a credit event
In a binary swap, the protection seller receives no recovery value in a credit event
DE R I VAT I VE
Client receives a higher spread to compensate for zero recovery in a credit event
CR E D IT
VAIDYA NATHAN 80
Digital (Binary) Default Swaps
Digital default swaps will demand a higher premium than a standard default swap
Its price will be sensitive to the recovery rate that has been assumed in the
calibration procedure
The higher the recovery rate in the calibration, the higher the calibrated hazard
rates will be
Since a digital default swap depends only on the hazard rates and not on the
recovery rate, it has a price that effectively increases with the assumed recovery
OP TI ON S
DE R I VAT I VE
CR E D IT
VAIDYA NATHAN 81
Digital Default Swaps
Based on run
above the quote
on ABY offers a
recovery bid-
offer of 35-45
with the
underlying ABY
5yr CDS spread at
297bp
An investor who
OP TI ON S
believes the
recovery rate on
ABY will be below
35 in the future
DE R I VAT I VE
should sell ABY

recovery at 35
CR E D IT
VAIDYA NATHAN 82
Digital Default Swaps
Executing this position will result in two trades:
Investor will be short protection on a DDS (Digital Default Swap) with a recovery
rate fixed at 35, a spread of 297bps, 5yr maturity
Investor will be long protection on a standard CDS, spread of 297bps, same maturity
date
The net of the two positions has zero carry
Profiting in Default:
If ABY defaults and bonds are trading at 20 (i.e. actual recovery is 20) investor
OP TI ON S
earns 15 points on the combined position:
On short protection position (DDS) investor loses 65 (100 – fixed recovery rate of 35)
DE R I VAT I VE
On the long protection position (regular CDS) investor earns 80 (100 less cost to buy
bond in market at 20)
CR E D IT
VAIDYA NATHAN 83
Calculate MTM on DDS
Calculate the MTM after one year assuming DDS is now trading at 20 instead of 35
Original positions:
Short fixed recover position at 35 (Loss on default is 100 – 35 = 65)
Long regular CDS position (Gain on default is 100 – R)
Unwind Positions:
Long fixed recovery position at 20 (Gain on default is 100 – 20 = 80)
Short regular CDS position (Loss on default is 100 – R)

OP TI ON S
Notional: USD 10 million, ignore discount rates
Survival
DE R I VAT I VE
Year Probability
2 0.95
3 0.90
CR E D IT
4 0.85
5 0.80 VAIDYA NATHAN 84
Cancelable Credit Swaps
Y bp per annum
Contingent Payment
(Par — Recovery)
Right to cancel
OP TI ON S
Seller receives a fee in return for making a Contingent Payment if there is a Credit Event of the
Reference Credit
Buyer decreases exposure to Reference Credit(s), but assumes contingent (“two-name”)

DE R I VAT I VE
exposure to Seller
Buyer has the right to cancel the trade (European or American option)
CR E D IT
VAIDYA NATHAN 85
Yield Pickup on structured CDS
Vanilla Mitsubishi Corp. CDS
3yrs @ 14bps, 5yrs @ 18bps, 10yrs @ 30bps
Cancelable Mitsubishi Corp. CDS
3yrs @ 15.5bps, 5yrs @ 20.5, 10yrs @ 33.5
Binary Mitsubishi Corp. CDS
3yrs @ 18.6bps, 5yrs @ 24bps, 10yrs @ 40bps

OP TI ON S
DE R I VAT I VE
CR E D IT
VAIDYA NATHAN 86
Outline
Page
Basket Products 26
Synthetic CDO 49
CDS Valuation 87

C R E D I T
VAIDYA NATHAN 87
Credit Spreads
Credit
Spreads
Credit Spreads
=
Default Probability
x (1-Recovery Rate)
Default Recovery
Probability Rate
VAL U A T I O N
C D S
VAIDYA NATHAN 88
Credit Default Swaps transfer default risk
X bp per annum
Protection Protection
Buyer Seller
Contingent Payment
(Par — Recovery)
Via a Credit Default Swap, the Protection Buyer transfers risk that the
Reference Entity will default
Protection Seller receives a fee, similar to an insurance premium, and

assumes the risk that the Reference Entity will default
Upon a default or Credit Event, protection seller makes a contingent

VAL U A T I O N
payment to the protection Buyer calculated according to the actual losses

(e.g. if Reference Entity recovers at 60%, protection seller pays 40%)
C D S
VAIDYA NATHAN 89
Caselet - EuroAutos
1.6% p.a., Q, Act/360

Broker/Dealer XYZ Bank
No Default Protection Seller Protection Buyer
Zero
Quarterly Payments Stop

Notional
Broker/Dealer XYZ Bank
Default Protection Seller Protection Buyer
Senior Unsecured Obligations with

notional amount of $10mn
VAL U A T I O N
C D S
VAIDYA NATHAN 90
Measure of default probability is required to price risky cash flow
Two sources exist:
Historical Analysis
Proprietary
i) Company/Sector expertise
Assumptions
Fundamental Analysis
ii) Market Data Calibrate probability against

market levels
Since we will be hedging with market instruments, it is essential that we

derive our default probabilities from the same instruments
As hedging instruments, credit default swaps offer superior liquidity,

lowest cost, and maximum flexibility
VAL U A T I O N
C D S
VAIDYA NATHAN 91
3
Specifications of credit markets are similar to interest rate markets
Yield
Market Details
Curve
Spd to
Currency USD 2
Mid
Trade Date 16-May-06 Swap Maturity Actual Mid Spreads Swap Zero
Days to Spot 2 # Maturity Dates Times yield Bid Mid Ask rates rates
Value Date 18-May-06 1 1D 19-May-06 0.003 3.500% 2.00 0.00 2.00 3.50% 3.56%
2 1M 19-Jun-06 0.088 3.700% 2.00 0.00 2.00 3.70% 3.76%
Conventions 3 2M 18-Jul-06 0.167 3.785% 2.00 0.00 2.00 3.79% 3.85%
Swap Basis 360 4 3M 18-Aug-06 0.252 3.870% 2.00 0.00 2.00 3.87% 3.93%
Swap Days B 5 4M 18-Sep-06 0.337 3.909% 2.00 0.00 2.00 3.91% 3.96%
MMkt DCC Act/365 6 5M 18-Oct-06 0.419 3.948% 2.00 0.00 2.00 3.95% 3.99%
Swap DCC Act/365 7 6M 20-Nov-06 0.510 3.987% 2.00 0.00 2.00 3.99% 4.03%
Swap cpns PA 2 8 9M 20-Feb-07 0.762 4.065% 2.00 0.00 2.00 4.07% 4.08%
Swap BDC M 9 1Y 18-May-07 1.000 4.130% 2.00 0.00 2.00 4.13% 4.13%
MMkt Basis 360 10 18M 19-Nov-07 1.507 4.177% 2.00 0.00 2.00 4.18% 4.22%
MMkt freq. 4 11 2Y 19-May-08 2.005 4.207% 2.00 0.00 2.00 4.21% 4.25%
VAL U A T I O N
Mean Reversion 0.00 12 3Y 18-May-09 3.003 4.254% 2.00 0.00 2.00 4.25% 4.30%
Interpolation type 1 13 4Y 18-May-10 4.003 4.284% 2.00 0.00 2.00 4.28% 4.33%
Libor Holidays GBP 14 5y 18-May-11 5.003 4.319% 2.00 0.00 2.00 4.32% 4.37%
C D S
VAIDYA NATHAN 92
3
Same as money market conventions
Market
Conventions
CCY AUD EUR GBP HKD JPY SGD USD USR

Days2Spot 1 1 1 1 1 1 1 1
Float Cnv Act/365F Act/360 Act/365F Act/365F Act/360 Act/365F Act/365 Act/360
Fixed Cnv Act/365F 30/360 Act/365F Act/365F Act/365F Act/365F Act/365 30/360
MM Basis 365 360 365 365 360 365 360 360
MM Freq 2 2 2 4 2 2 4 4
Swap CPA 2 1 2 4 2 2 2 2
Swap Days A B A A A A B B
Swap Basis 365 360 365 365 365 365 360 360
Swap Bad Day

Conv M M M M M M M M
LIBOR Hol AUD EUR GBP HKD JPY SGD GBP USR
VAL U A T I O N
C D S
VAIDYA NATHAN 93
3
Additional aspects
Reference Details Credit
Reference
Currency USD Parameters
Trade Date 16-May-06 Fee Pay Payment Payment Mean Spread Recovery
Days to Spot 3 Interval Accrued Index Day Count Bad Day Reversion Interpolation Rate
Settlement Date 19-May-06 Convention Convention (%)
Q FALSE Q Act/365 M 0.00 1 50.00%
Credit Spreads
Tenors 1D 1M 3M 6M 1Y 2Y 3Y
Dates 22-May-06 19-Jun-06 21-Aug-06 20-Nov-06 21-May-07 19-May-08 19-May-09
Bid-Offer
3.0 3.0 3.0 3.0 3.0 3.0 3.0
Spread
Bid Spread 98.50 98.50 118.50 138.50 148.50
Mid Spread 100.0 100.0 120.0 140.0 150.0
Ask Spread 101.50 101.50 121.50 141.50 151.50

Clean Spreads 2.005% 2.005% 2.005% 2.005% 2.417% 2.836% 3.049%
Duration 0.254 0.496 0.967 1.860 2.689
Credit
Reference
VAL U A T I O N
Volatility
Structure
Tenors 1D 1M 3M 6M 1Y 2Y 3Y
Dates 20-May-06 19-Jun-06 19-Aug-06 19-Nov-06 19-May-07 19-May-08 19-May-09
ATM Volatilities 15.00% 15.00% 15.00% 15.00%
C D S
VAIDYA NATHAN 94
3
Trade Specifications
Trade Summary CDS References
CDS Price (Deterministic) 372,020.31 CDS Fee 120.00 bp
CDS Price (Tree) 360,488.06 Trade Recovery Rate 50.00%
Opt. Premium (Tree) 0.00 Maturity 2Y
Opt. Premium(Deterministic) 0.00 Coupon Interval Q
Hedge Cost 0.00 First Fixing Date 19-May-2006
Total 360,488.06 Next Regular Fixing Date 19-May-2006
Trade Inputs Maturity Date 19-May-2008
Trade Notional 100,000,000.00 Fee Day Count ACT/365
Trade Date 16-May-2006 Accrual Bad Day M
Sttlement Date 19-May-2006 Payment Bad Day M
Forward Start Time(eg. 1M, 1Y) 1y CDS Structure
Call / Put Call Pay Accrued Fee on Default FALSE
CDS (Long / Short) Long Pay at Maturity FALSE
Option (Long / Short) Long CDS Fee payment tlll maturity FALSE
Option References Option Exercise Schedule

VAL U A T I O N
Option Maturity 1Y Exercise Exercise Strike
Exercise Interval Q # Unadjusted Adjusted Date(s)

Exercise Effective
19-May-2006
Date
C D S
Maturity Date 21-May-2007 0 19-May-06 19-May-06 19-May-06

VAIDYA NATHAN 95
3
CDS Cashflow due from Buyer and Seller of Protection
Buyer of protection pays quarterly premium to seller until the earlier of a credit event or maturity
X X X X X
3m 6m 9m 12m 15m 18m ... 60m
At inception: PV of both legs are equal

Credit Event
3m 6m 9m 12m 15m 18m ... 60m
100 - R
VAL U A T I O N
The seller of protection pays par less recovery to the protection buyer if there is a credit event
during the life of the contract
C D S
VAIDYA NATHAN 96
CDS Cashflow due from Buyer and Seller of Protection
N
Risky PVFIXED = ∑ S .DFi .SPi .α i
i =1
S is the per-annum CDS spread
N is the number of coupon periods
DFi is the riskless discount factor from time t0 to ti
SPi is the Survival Probability of the reference entity from time t0 to ti
αi is the accrual factor from ti-1 to ti
R is the recovery rate on the delivered obligation

N
Risky PVFLOATING = ∑ (1 − R ).DFi .(SPi −1 − SPi )
VAL U A T I O N
i =1
C D S
VAIDYA NATHAN 97
Valuation of any risky cash flow is based on concept of risky PV
As default risk increases, the PV of a risky cash flow decreases
This corresponds to discounting a risky cash flow with a risky discount

factor
A risky discount factor is alternatively expressed as the product of a
risk-free discount factor and a survival probability
To price a “risky” instrument we therefore need

Payment structure
Risk free discount curve
Default probability curve
VAL U A T I O N
C D S
VAIDYA NATHAN 98
2
Deriving default probability from CDS spreads
A
A 1-year
1-year default
default swap
swap with
with annual
annual coupon
coupon
1 year annual CDS

P ND
( 0 ,1) S
(1 − P( ND
0 ,1 ) )
-(1-R)
Period 1
1 ⎡P ND x S - (1 − P ND ) x (1 - R)⎤
0=
(1 + r ) ⎢⎣ ⎥⎦
risk free
VAL U A T I O N
C D S
VAIDYA NATHAN 99
4
By bootstrapping, a term structure of default prob can be estimated
Default
Default probability
probability tree
tree construction
construction
P( 1ND
,2 )
S
S
ND
P ( 0 ,1)
0 (1 − P( 1ND
,2 ) )
(100-R)
(1 − P( ND
0 ,1 ) ) (100-R)
Period 1 Period 2
Using bootstrapping method we can use P(0ND

,1) solved in Period 1 to deduce P(1ND
,2)
in Period 2
VAL U A T I O N
C D S
V A I D Y A N A T H A N 100
5
Useful Rule of Thumb
Market equilibrium should ensure that expected loss is equal to the PV of

any spread paid in compensation for bearing the risk:
Spread Expected
loss
S = P D (1 − R)
Rearranging gives a simple expression for default probability in terms of
CDS spread and recovery rate
S
P =D
(1 − R)
VAL U A T I O N
1
1−
p(t,T ) =
(1 + s(t : t,T )n )
1− θ
C D S
6
Calculation of default prob is complicated by a number of factors
Recovery assumptions
Recovery amount cannot be known in advance; therefore an assumption
must be made
Assumptions about what will be recovered can vary:
— Recover a flat cash amount
— Recover a percentage of risk free PV
— Recover a percentage of outstanding notional plus accrued interest
Interpolation
Credit default swaps typically pay fee quarterly
Fee accrual
In a standard CDS, only the accrued spread is usually paid at time of
default
VAL U A T I O N
C D S
7
Effect of recovery assumption on implied survival probability
Higher Default
Probability
High Recovery
Lower Survival
Assumption
Probability
Higher Survival
Low Recovery Probability
Assumption
Lower Default
Probability
VAL U A T I O N
C D S
Survival probability with time for different recovery rates
Survival
Survival probability
probability with
with time
time for
for different
different recovery
recovery rates
rates (CDS
(CDS Spread
Spread =
= 100
100 bps)
bps)
100.00%
Survival Probabilities with Time
90.00%
80.00%
70.00%
60.00%
50.00%
R= 90% R= 50% R= 10%
40.00%
30.00%
20.00%
10.00%
Time (in y rs)

0.00%
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
VAL U A T I O N
For low recovery rate assumptions, survival probability decreases approximately linearly
over time. For high recovery rate assumptions, this relationship is more ‘convex’
C D S
Decline in survival probability with higher recovery rates
Survival
Survival probability
probability with
with time
time for
for different
different recovery
recovery rates
rates (CDS
(CDS Spread
Spread =
= 100
100 bps)
bps)
Maturity R= 90% R= 50% R= 10%
0 99.99% 100.00% 100.00%
1 91.12% 98.09% 98.93%
2 83.02% 96.21% 97.87%
3 75.64% 94.37% 96.82%
4.0027 68.92% 92.56% 95.78%
5.0027 62.80% 90.79% 94.75%
6.0027 57.21% 89.05% 93.74%
7.0027 52.13% 87.35% 92.73%
8.0055 47.50% 85.67% 91.73%
9.0055 43.28% 84.04% 90.75%
10.0055 39.43% 82.43% 89.78%
11.0055 35.93% 80.85% 88.82%
12.0082 32.73% 79.30% 87.86%
13.0082 29.82% 77.78% 86.92%
14.0082 27.17% 76.29% 85.99%
15.0082 24.76% 74.83% 85.07%
16.011 22.55% 73.40% 84.15%
17.011 20.55% 72.00% 83.25%
18.011 18.73% 70.62% 82.36%
19.011 17.06% 69.27% 81.48%
20.0137 15.55% 67.94% 80.60%
VAL U A T I O N
C D S
Caselet: a typical problem of front & middle office folks
Bid Mid Ask

Client entered into the following CDS trade a year back Tenors Spread Spread Spread
Reference Entity: AT&T Corporation 1D 98.5 100 101.5
1M 98.5 100 101.5
Maturity: 5 years
2M 98.5 100 101.5
Notional: USD 10 million
3M 98.5 100 101.5
Contract Spread: 150 bps 4M 98.5 100 101.5
Current bid/offer for AT&T is as below 5M 98.5 100 101.5
6M 98.5 100 101.5
Calculate the MTM on the trade
9M 98.5 100 101.5
1Y 98.5 100 101.5
18M 98.5 100 101.5
2Y 98.5 100 101.5
VAL U A T I O N
3Y 98.5 100 101.5

4Y 98.5 100 101.5
5Y 98.5 100 101.5
C D S
Survival Probabilities As Weighting Factors
K
MTM = ∑ (Current CDS − ContractCDS).DFi .SPi
i =1
Current CDS is the CDS spread currently prevailing in the market
Contract CDS is the CDS spread at which the trade was entered
K is the number of coupon periods
DFi is the riskless discount factor from time t0 to ti
SPi is the Survival Probability of the reference entity from time t0 to ti

VAL U A T I O N
C D S
Conceptualising CDS Mark-to-Market
ORIGINAL
X X X X X X X
NO CREDIT EVENT
3m 6m 9m 12m 15m 18m ... 60m
X-Y X-Y X-Y
=
15m 18m ... 60m
OFFSET
3m 6m 9m 12m 15m 18m ... 60m
VAL U A T I O N
Y Y Y
C D S
But the Cash Flows are risky …
ORIGINAL
X X X X X
Nominal Value
CREDIT EVENT
3m 6m 9m 12m 15m 18m ... 60m
Defaulted Obligation X-Y X-Y X-Y
Credit Event
=
15m 18m ... 60m
OFFSET Defaulted Obligation Annuity Cancelled
3m 6m 9m 12m 15m 18m ... 60m Default Payments Net off

VAL U A T I O N
Nominal Value
Y
C D S
Forward default probability for different recovery rates
Forward
Forward default
default probability
probability with
with time
time for
for different
different recovery
recovery rates
rates (CDS
(CDS Spread
Spread =
= 100
100 bps)
bps)
Forward Default Probabilities with Time
10.00%
9.00%
8.00%
7.00%
6.00%
5.00%
R= 90% R= 50% R= 10%
4.00%
3.00%
2.00%
1.00%
0.00% Time (in yrs)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
VAL U A T I O N
For low recovery rate assumptions, forward default probability decreases approximately
linearly over time. For high recovery rate assumptions, it decreases exponentially
C D S
Increased forward default probability with higher R
Forward
Forward default
default probability
probability with
with time
time for
for different
different recovery
recovery rates
rates (CDS
(CDS Spread
Spread =
= 100
100 bps)
bps)
Maturity R= 90% R= 50% R= 10%
1 8.87% 1.91% 1.07%
2 8.10% 1.88% 1.06%
3 7.38% 1.84% 1.05%
4.0027 6.72% 1.81% 1.04%
5.0027 6.12% 1.77% 1.03%
6.0027 5.58% 1.74% 1.02%
7.0027 5.08% 1.70% 1.01%
8.0055 4.63% 1.67% 0.99%
9.0055 4.22% 1.64% 0.98%
10.0055 3.85% 1.61% 0.97%
11.0055 3.50% 1.58% 0.96%
12.0082 3.20% 1.55% 0.95%
13.0082 2.91% 1.52% 0.94%
14.0082 2.65% 1.49% 0.93%
15.0082 2.41% 1.46% 0.92%
16.011 2.20% 1.43% 0.91%
17.011 2.00% 1.40% 0.90%
18.011 1.82% 1.38% 0.89%
VAL U A T I O N
19.011 1.66% 1.35% 0.88%

20.0137 1.52% 1.33% 0.88%
C D S
Issuer-Weighted Recovery Rate Descriptive Statistics
Mean (1982 -
Priority in Capital Structure 2003) Mean (2004)
Senior Secured 57.4% 80.8%
Senior Unsecured 44.9% 50.1%
Senior Subordinated 39.1% 44.4%
Subordinated 32.0% NA
Junior Subordinated 28.9% NA
All Bonds 42.2% 54.3%
VAL U A T I O N
C D S
Effect of recovery assumption on risky duration
Higher Default
Probability
High Recovery Lower Survival
Assumption Probability
Lower “Risky
Duration”
Higher “Risky
Duration”
Higher Survival
Low Recovery Probability
Assumption
VAL U A T I O N
Lower Default
Probability
C D S
Risky Duration for different credit spreads
Risky
Risky Duration
Duration for
for constant
constant term
term structure
structure of
of credit
credit spreads
spreads
8
Risky duration for different credit spreads
Spread 50 bps flat Spread 100 bps flat Spread 200 bps flat
2
VAL U A T I O N
Maturity
1D 1M 2M 3M 4M 5M 6M 9M 1Y 18M 2Y 3Y 4Y 5Y 6Y 7Y 8Y 9Y 10Y
C D S
Nonlinearity of risky duration
Nonlinearity
Nonlinearity of
of risky
risky duration
duration for
for half
half and
and double
double credit
credit spreads
spreads
Spread 50 Spread 100 Spread 200
Maturity Mat times bps flat Duration bps flat Duration bps flat Duration
1D 0.003 50.0 0.0027 100.0 0.0027 200.0 0.0027
1M 0.077 50.0 0.0764 100.0 0.0764 200.0 0.0763
2M 0.170 50.0 0.1684 100.0 0.1681 200.0 0.1675
3M 0.244 50.0 0.2410 100.0 0.2404 200.0 0.2392
4M 0.329 50.0 0.3245 100.0 0.3237 200.0 0.3220
5M 0.416 50.0 0.4100 100.0 0.4087 200.0 0.4061
6M 0.496 50.0 0.4869 100.0 0.4851 200.0 0.4816
9M 0.748 50.0 0.7297 100.0 0.7261 200.0 0.7190
1Y 1.000 50.0 0.9693 100.0 0.9634 200.0 0.9517
18M 1.496 50.0 1.4316 100.0 1.4194 200.0 1.3955
2Y 2.000 50.0 1.8893 100.0 1.8687 200.0 1.8285
3Y 3.008 50.0 2.7721 100.0 2.7330 200.0 2.6604
4Y 4.005 50.0 3.5954 100.0 3.5253 200.0 3.3929
5Y 5.003 50.0 4.3756 100.0 4.2687 200.0 4.0663
6Y 6.003 50.0 5.1179 100.0 4.9708 200.0 4.6947
7Y 7.005 50.0 5.8222 100.0 5.6303 200.0 5.2733
8Y 8.014 50.0 6.4956 100.0 6.2596 200.0 5.8261
VAL U A T I O N
9Y 9.011 50.0 7.1196 100.0 6.8308 200.0 6.3039

10Y 10.008 50.0 7.7092 100.0 7.3654 200.0 6.7428
C D S
Outline
Page
Basket Products 26
Synthetic CDO 49
CDS Valuation 87

C R E D I T
Framework
Every credit default swap is documented in a contract based on the ISDA format -
called the Confirmation
The terms used in the Confirmation are defined in the 2003 Credit Derivatives
Definitions (formerly 1999 Definitions)
A high level of standardisation of documentation exists in the market
Standardization makes credit default swaps easier to trade, creates transparency

and facilitates market participation
R O ADMAP
DE R I VAT I VE S
CR E D IT
Key Contract Terms
Reference Entity - the entity that credit protection covers
Obligations - Borrowed Money, Bonds or Loans are types of obligations that the
protection covers
Credit Events - the triggers are Bankruptcy, Repudiation/Moratorium, Failure to

Pay and Restructuring, Obligation Acceleration
Bankruptcy – covers insolvency, appointment of administrators/liquidators, creditor
arrangements, etc
Failure to Pay – on one or more Obligations after expiration of any applicable grace period
Restructuring – agreement between Reference Entity and holders of any Obligation (and
R O ADMAP
such agreement is not provided for under the terms of that Obligation) with respect to
reduction of interest or principal, postponement of payment of interest or principal, change
of currency (other than “Permitted Currency”) and subordination
DE R I VAT I VE S
Deliverable Obligations - settle contracts with Bonds or Loans with predefined

characteristics
CR E D IT
Key Contract Terms - Deliverable Obligations
If a Credit Event Deliverable Obligation Deliverable Obligation

occurs, the Buyer of Categories: Characteristics:
protection can
No Payment Yes Not Subordinated
deliver Deliverable Specified Currency -
No Borrowed Money Yes
Obligations to the Standard Specified Currencies
Reference
Seller No Obligation(s) Only No Not Sovereign Lender
No Not Domestic Currency

No Bond
Deliverable No Not Domestic Law
No Loan No Listed
Obligations are not Yes Bond or Loan Yes Not Contingent
the same as No Not Domestic Issuance
Yes Assignable Loan
Obligations — they Yes Consent Required Loan
R O ADMAP
are more narrowly No Direct Loan Participation

No Indirect Loan Participation
defined Qualifying Participation Seller
Yes Transferable
DE R I VAT I VE S
30 years Maximum Maturity

No Accelerated or Matured
Yes Not Bearer
CR E D IT
2003 Definitions - Why Introduce and Key Changes
Why Introduce
To consolidate market experience - the 2003 Definitions represent a development
of the 1999 Definitions.
Too many supplements (now incorporated)
Modified Restructuring was not adopted in Europe
Time to overhaul and clean up definitions
Key Changes
Modified Modified Restructuring for Europe
R O ADMAP
New Settlement Fallbacks

New Guarantee provisions
DE R I VAT I VE S
CR E D IT
CDS Structural Roadmap
REFERENCE ENTITY Underlying credit risk transferred?
CREDIT EVENT Types of "default" covered
OBLIGATIONS Default on which instruments qualify
PROTECTION PERIOD Occurance of Credit Event

R O ADMAP
REFERENCE OBLIGATION Seniority of exposure transferred

DE R I VAT I VE S
DELIVERABLE OBLIGATION Instruments used for settlement
PHYSICAL SETTLEMENT CDS settlement

CR E D IT
Caselet: Armstrong World Industries
US
US company
company Armstrong
Armstrong World
World Industries
Industries missed
missed payments
payments on
on its
its debt
debt
US company Armstrong World Industries missed payments on its debt,

which triggered credit default swaps
Its parent company Armstrong Holdings however, did not default
Many market participants had treated the parent and principal subsidiary
interchangeably and had hedged positions with offsetting contracts in the
other entity
The lesson here is that there may be substantial credit basis risk between
different entities in the same group
R O ADMAP
Worse still, certain contracts in the market had referenced simply

Armstrong without clarifying to which specific entity the contract
DE R I VAT I VE S
referred
CR E D IT
Caselet: National Power
National
National Power
Power PLC
PLC demerged
demerged certain
certain assets
assets and
and subsidiaries
subsidiaries into
into two
two entities
entities
In November 2000, National Power PLC of the UK demerged certain assets

and subsidiaries into two entities: Innogy and International Power
In consideration for the transfer of assets to Innogy, shareholders were

given holdings in the new entity
National Power then changed its name to International Power
Innogy also assumed certain debt obligations of National Power
This demerger prompted substantial debate as to whether Innogy had

R O ADMAP
become a Successor
DE R I VAT I VE S
CR E D IT
Non Sovereign Decision Tree
Non-Sovereign
Non-Sovereign Successor
Successor Summary
Summary Decision
Decision Tree
Tree
Does New Entity have >75% of It is the Sole Successor for the for the Entire
Obligations? Credit Derivative Transaction
YES
NO
YES YES
Does New Entity have <25% of Does Reference No Change to
R O ADMAP
Obligations? Entity still exist Contract
NO
DE R I VAT I VE S
NO
New Entity Taking
Each Successor Assigned New Largest % of
Credit Derivative Transaction – Relevant Obligations
Includes Reference Entity >25% is Successor
CR E D IT
Caselet: Xerox Corporation
Xerox
Xerox extended
extended the
the date
date for
for repayment
repayment of
of principal
principal
In the summer of 2002, as part of a wider agreement with its banks,

Xerox extended the date for repayment of principal
This was in respect of a major syndicated bank facility that was due for
repayment in September
However, market participants entered a legal dispute about whether this

was a result of a deterioration in creditworthiness
And over what period prior to Restructuring such deterioration could

reasonably have occurred
R O ADMAP
DE R I VAT I VE S
CR E D IT
Caselet: Argentina
Obligation
Obligation Exchange
Exchange requirements
requirements
Obligation Exchange requirements became the subject of legal disputes
Argentina was facing a tight liquidity situation
It “requested” local investors to exchange $50bn of bonds for new issues

with lower coupons
In question was the meaning of “mandatory” in such circumstances

R O ADMAP
DE R I VAT I VE S
CR E D IT
Caselet: Railtrack
Bankruptcy
Bankruptcy Credit
Credit Event
Event
On 7 October 2000, Railtrack plc was placed by the UK government into

Special Railways Administration
This constituted a Bankruptcy Credit Event
The announcement date was a Saturday
Investors who bought credit default swap protection on the Wednesday,

Thursday or Friday of the previous week would have not been covered for
this Credit Event
R O ADMAP
Under the current conventions, however, such risks are considerably

reduced
DE R I VAT I VE S
CR E D IT
Caselet: Railtrack CTD Obligation
“Widows
“Widows and
and orphans”
orphans” clause
clause
Following the Railtrack Bankruptcy Credit Event in 2000, the cheapest-to-

deliver obligation was the 3.5% of 2009 exchangeable bond
Most of the market took the view that, provided the bond is
exchangeable or convertible at the option of the holder, the bondholder
should be the beneficiary and the exchange or conversion option within
its control
One further complication in the Railtrack case was the inclusion of a so

called “widows and orphans” clause in the exchangeable bond which gave
R O ADMAP
the trustee the right to force conversion of the bond on the holder in
certain circumstances where it was viewed as being in the interests of
the investor
DE R I VAT I VE S
After a protracted legal dispute, in February 2003, UK courts ruled in

favour of deliverability
CR E D IT
Caselet: Marconi
Somewhat
Somewhat unusual
unusual guarantee
guarantee structure
structure
The Marconi group had a somewhat unusual guarantee structure
The holding company Marconi PLC provided lenders and bondholders of

subsidiary Marconi Corporation PLC with a guarantee
Although the bond guarantees were stated to be “unconditional” they

contained a provision that they would fall away upon the repayment of
certain other guaranteed obligations
In 2002 a Bankruptcy Credit Event occurred in relation to Marconi, and

the approach of market participants was to deliver loans instead of
R O ADMAP
bonds, so as to avoid the risk that the guarantee structure would render
the bonds undeliverable (under 1999 Definitions)
The main exception to this, was where the bond in question was stated as
DE R I VAT I VE S
the Reference Obligation since in most circumstances this is deliverable

CR E D IT
Caselet: Xerox Syndicated Bank Loan extension
Pressure
Pressure on
on Mod-R
Mod-R
Mod-R worked pretty well in the US till it came under pressure
In summer 2002, Xerox extended maturities of a syndicated bank loan
In this case the maturity limitation requirements of Mod-R did not really
insulate Sellers of protection from the “cheapest-to-deliver” risk
This was because, although not long dated, Xerox’s yen bonds were
trading about 15-20 points lower than where the dollar bank loans were
quoted
R O ADMAP
DE R I VAT I VE S
CR E D IT
Modified Modified Restructuring
New features:
The Restructured Obligation must be a Multiple Holder Obligation (i.e.

more than three holders)
If Buyer triggers the contract
— Deliverable Obligations subject to a maturity cap of 60 months for the

Restructured Bond or Loan, 30 months for others AND must be
R O ADMAP
Conditionally Transferable Obligations (matches LMA standard)
— Buyer may elect to partially settle

DE R I VAT I VE S
CR E D IT
Modified Modified Restructuring
In 2005, a customer buys 5 year protection on EnergyCo
June 2006, EnergyCo enters legally binding agreement to restructure

certain of its bonds
You can deliver (1) restructured bonds maturing before mid 2011 and
(2) non restructured bonds maturing before 2010
30 month 60 month
Maturity Maturity Maturity
Cap Floor Cap
R O ADMAP
DE R I VAT I VE S
2005 2006 2007 2008 2009 2010 2011 2012

Effective Legally effective STD
date of
Restructuring
CR E D IT
Modified Modified Restructuring - US and Europe
Feature: MR (US standard) MMR (European Standard)
Maturity Cap: 30 month cap • 30 month cap for non

Floored at STD restructured obligations
• 60 month cap for restructured
obligations
• Floored at STD
Transferability: Must be transferable to Must be transferable to entities

extensive list of entities regularly engaged in loan and
without consent. securities markets with consent
not to be unreasonably withheld
R O ADMAP
Obligations covered: At least three holders and At least three holders and in the
DE R I VAT I VE S
requires a 2/3 majority to case of a Loan, requires a 2/3

implement restructuring. majority to implement
restructuring
CR E D IT
New Settlement Fallbacks
To avoid failed contracts, parties now have an indefinite period of time to settle
the contract
Buyer must attempt scheduled settlement but if this fails, fallbacks will apply
Buyer may continue to attempt delivery
Seller may close out by buying in the Deliverable Obligation or nominating an

alternative for delivery
R O ADMAP
DE R I VAT I VE S
CR E D IT
New Settlement Fallbacks - Bond Delivery
Buyer has 30 calendar days to deliver a Settlement Notice
Buyer has 30 Business Days + 5 Business Day fallback to effect delivery of the Bonds
If Delivery has not occurred by this date, Seller may buy the Bond in at the lowest
offer. Seller has 4 Business Days to complete the process
If Seller fails to complete the process, Buyer may continue to attempt delivery.
Buyer can not deliver whilst the buy-in process is in operation
Seller may try to buy the Bond in again, but must wait at least 8 Business Days
before restarting the process
R O ADMAP
DE R I VAT I VE S
CR E D IT
New Settlement Fallbacks - Bond Delivery
Buy-in may start any time
Buy-in max Buy-in max

4 BD ≥ 8 BD 4 BD
30 days 30 + 5 BD
Event Last day to deliver Last day for

Determination Notice of Physical delivery of Bonds
R O ADMAP
Date Settlement before fallbacks

begin
Buyer may continue to
attempt to deliver between
DE R I VAT I VE S
buy-in attempts by Seller

CR E D IT
New Settlement Fallbacks - Loan Delivery
Buyer has 30 calendar days to deliver a Settlement Notice
Buyer has 30 Business Days + 5 Business Day fallback to effect delivery of the Loan
If Delivery has not occurred by this date, Buyer may deliver a Transferable Bond
or an Assignable Loan instead provided that Buyer provides certification from a
Managing Director that reasonable efforts were used to get consent
If Buyer has not delivered anything for a further 15 Business Days, Seller may
nominate an Assignable Loan or a Transferable Bond and require Buyer to purchase
and deliver
R O ADMAP
DE R I VAT I VE S
CR E D IT
New Settlement Fallbacks - Loan Delivery
Buyer can deliver any other Assignable

Loan or Transferable Bond
Seller can nominate an available

30 days 30 + 5 BD 15 BD Loan or Bond
R O ADMAP
Event Last day to deliver Last day for delivery

Determination Notice of Physical of loans before
Date Settlement fallbacks begin
DE R I VAT I VE S
CR E D IT
New Guarantee Provisions
Users can now select what type of guarantees can trigger the contract and what is
deliverable
Guarantee has to be a “Qualifying Guarantee” i.e. a written instrument where

Reference Entity irrevocably agrees to make payment
Upstream, downstream, side-stream and third party guarantees may be identified

and treated separately
Europe “All Guarantees” will be adopted. In US “Qualifying Affiliate Guarantees”

will be adopted
Qualifying Affiliate Guarantee – Reference Entity guarantees debt of an affiliate

R O ADMAP
where it owns more than 50 percent of the voting shares of that affiliate
DE R I VAT I VE S
CR E D IT
New Guarantee Provisions
Europe and Asia –
covers all these
Qualifying Guarantees
Parent
Company
50%
50% Guarantee
Operating Reference
Company Guarantee Entity Guarantee
US - the only Qualifying Third Party

R O ADMAP
Guarantee
Affiliate Guarantee – 50%
downstream with a
holding of at least 50% Subsidiary
DE R I VAT I VE S
CR E D IT
Other Key Changes
CHF is now a standard deliverable currency
Notice of Intended Physical Settlement becomes Notice of Physical Settlement

(“NPS”)
Pari Passu Ranking becomes Not Subordinated
Minor amendments to Restructuring definition, Successor definitions
Not Contingent definition revised to remove need for a coupon
Convertibles language broadened to accommodate wider range of deliverables
Public Sources broadened to include Australian Financial Review et al plus the main
R O ADMAP
source of business news in country of Reference Entity
Scheduled Termination Date no longer subject to adjustment

DE R I VAT I VE S
CR E D IT
Other Key Changes
Modified Following replaced with Following as standard convention for all trades
Valuation provisions restructured so that all Firm Bids are used regardless of
Quotation Size, with zero only deemed for the no bid portion
Repudiation/Moratorium redrafted to require a subsequent Failure to Pay (in any

size) before the earlier of (a) 60 days and (b) the next payment date for the
instrument. Buyer must also deliver a notice of a Potential Repudiation/Moratorium
R O ADMAP
DE R I VAT I VE S
CR E D IT
Outline
Page
Basket Products 26
Synthetic CDO 49
CDS Valuation 87

C R E D I T
Applications for credit derivatives in the global market
Motivations
Motivations for
for using
using Credit
Credit Derivatives
Derivatives
1. Trading/ market making
2. Product structuring
3. Hedging trading
instruments
4. Active portfolio/ asset

management
U PDAT E
5. Management of
individual credit lines
DE R I VAT I VE S
6. Management of
regulatory capital
7. Management of
economic capital
CR E D IT
Rankings for application of Credit Derivatives
Applications for credit derivatives Rankings 2003 Rankings 2006 (E)
Trading/market making 1 1
Product structuring 2 2
Hedging trading instruments 3 4
Active portfolio/asset management 4 3
Management of individual credit lines 5 5

U PDAT E
Management of regulatory capital 6 7

DE R I VAT I VE S
Management of economic capital 7 6

CR E D IT
Credit Derivatives positions
Credit
Credit Derivatives
Derivatives positions
positions
1997 180
1998 350
1999 586
2000 893
2001 1189
2002
U PDAT E
1952
2003 3548
DE R I VAT I VE S
2004 5021
2006 (E) 8206

CR E D IT
Comparative Interest rate derivatives growth
Interest
Interest Rate
Rate Growth
Growth (USD
(USD billion)
billion)
141,991
121,799
101,658
89,955
77,568
67,465
64,125 64,668
60,091
54,072
50,015
42,368
U PDAT E
DE R I VAT I VE S
Jun-98 Dec-98 Jun-99 Dec-99 Jun-00 Dec-00 Jun-01 Dec-01 Jun-02 Dec-02 Jun-03 Dec-03
CR E D IT
Breakdown of market participation
Market
Market Composition
Composition
24%
42%
34%
U PDAT E
DE R I VAT I VE S
Intermediary / Market Maker Buyer Seller

CR E D IT
Institutions using credit derivatives to buy protection
Buyers
Buyers of
of credit
credit protection
protection
3% 3% 1%
2006 (Expected) Other 2003
7% Banks
Pension fundsAgencies 3% Securities houses
2%
Mutual funds 5%
Hedge funds
5%
Corporates
Insurance 16%
51% Insurance Companies
Companies
9% Mutual funds
Pension funds
Banks
Corporates 43% Other Agencies
4% 16%
1% 1% 1%
1999 7%
U PDAT E
6% Banks
Securities houses
3%
Hedge funds Hedge funds
DE R I VAT I VE S
17% Corporates
Insurance Companies
18% Mutual funds

Securities Pension funds
63%
houses Other Agencies
CR E D IT
15%
Sellers of credit protection
Institutions
Institutions using
using credit
credit derivatives
derivatives to
to sell
sell protection
protection
4% 1%
4%
2006 (Expected) 2003
Other Banks
Pension fundsAgencies Securities houses

1% Hedge funds
Mutual funds 6%
6% 20% 38% Corporates
Insurance Companies
Banks 2% Mutual funds
34% Pension funds
Other Agencies
Insurance 15%
Companies 16%
21% 1999
2% 3% 1%
U PDAT E
Banks
Securities houses
Hedge funds
DE R I VAT I VE S
Corporates 23% Corporates

3% Insurance Companies
47%
Securities Mutual funds
Hedge funds 3%
houses
15% Pension funds
14% 5%
CR E D IT
Other Agencies
16%
Credit Derivatives by region
Credit
Credit Derivatives
Derivatives by
by region
region in
in 2006
2006 (Expected)
(Expected)
3%
39% 43%
U PDAT E
10% 5%
DE R I VAT I VE S
CR E D IT
London Europe ex-London Asia/Australia US Other

Credit Derivatives by region in 2003 & 1999
Credit
Credit Derivatives
Derivatives by
by region
region
2003 1999
45%
44%
41%
40%
U PDAT E
DE R I VAT I VE S
9% 8%
6% 5%
1% 1%
CR E D IT
London Europe ex- Asia/Australia US Other London Europe ex- Asia/Australia US Other
London London
Credit Derivatives booked by region
Credit
Credit Derivatives
Derivatives booked
booked by
by region
region
14%
6%
14%
66%
U PDAT E
DE R I VAT I VE S
London Europe ex-London Asia/Australia US

CR E D IT
Credit Derivatives Market Size by region
Credit Derivatives Market Size (US$ bn) 2003 2004 2006
Global market size 3,548 5,021 8,206
London market size 1,586 2,230 3,563
Americas market size 1,459 2,000 3,173
Asia/Australia market size 287 446 858

U PDAT E
Other Europe/Rest of World 216 345 612

DE R I VAT I VE S
CR E D IT
Global Credit Derivatives Product Usage
Global
Global Credit
Credit Derivatives
Derivatives Product
Product Usage
Usage
2% 1% 1%
4%
4% 51%
Single-name credit default swaps
4%
Synthetic CDOs – full capital
Synthetic CDOs – partial capital
6%
Full index trades
Tranched index trades
2%
Credit linked notes
Total return swaps
U PDAT E
9% Basket products
Asset swaps
DE R I VAT I VE S
Credit spread options

Swaptions
10% Equity linked credit products
CR E D IT
6%
Current Product Usage
Global
Global Credit
Credit Derivatives
Derivatives Product
Product Usage
Usage –– 2006
2006 (Expected)
(Expected)
3%
3% 1%
3%
42%
5%
Single-name credit default swaps
4% Synthetic CDOs – full capital
Synthetic CDOs – partial capital
Full index trades
6%
Tranched index trades
Credit linked notes
Total return swaps
5%
U PDAT E
Basket products
Asset swaps
DE R I VAT I VE S
Credit spread options

Swaptions
12%
Equity linked credit products
5%
11%
CR E D IT
Underlying reference entity
Category
Category of
of underlying
underlying reference
reference entity
entity
2006 (Expected) 2003 2%

5%
Corporate assets
7%
4% 2%
Financials
8%
22% Sovereign assets

(emerging markets)
64% Sovereign assets (non-
emerging markets)
22% Other
64% 1999
3%
U PDAT E
6% Corporate assets
9%
Financials
DE R I VAT I VE S
Sovereign assets
Corporate assets 60% (emerging markets)
22%
Financials Sovereign assets (non-
Sovereign assets (emerging markets) emerging markets)
CR E D IT
Sovereign assets (non-emerging markets) Other
Other
Credit rating of underlying reference entity
Credit
Credit rating
rating of
of the
the underlying
underlying reference
reference entity
entity
2006 (Expected) 2003 AAA – AA

AAA – AA
19%
65%
3%
Below B A – BBB
16%
13%
4% 63%
Below B A – BBB
1999 BB – B
17% AAA – AA
U PDAT E
DE R I VAT I VE S
17%
66%
0%
BB – B Below B A – BBB
17%
CR E D IT
BB – B
Tenor distribution
Maturity
Maturity
2006 (Expected) 2003 2% 7%

Over 10 years
2% 16%
Under 1 year
Under 1 year
1 – 5 years
7%
5 years
5 – 10 years 21% 5 – 10 years
21% 54% Over 10 years
1999 5% 9%
U PDAT E
9%
Under 1 year
5 years 1 – 5 years
1 – 5 years
DE R I VAT I VE S
18% 52%
5 years
41% 5 – 10 years
36%
Over 10 years
CR E D IT
Market Constraints
Constraints
Constraints in
in using
using Credit
Credit Derivatives
Derivatives
1. Lack of client
knowledge of the product
2. Regulatory constraints
3. Systems / Infrastructure
4. Pricing – lack of data

U PDAT E
5. Lack of agreed
accounting conventions
DE R I VAT I VE S
6. Lack of homogenous
documentation
7. Lack of market liquidity

and depth
CR E D IT

Credit Derivatives Overview IIMA

Загружено:

Сведения о документе

Исходное описание:

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Credit Derivatives Overview IIMA

Загружено:

Авторское право:

Доступные форматы

JULY 2006

Credit Derivatives Overview 1

CLN & Linear Basket 14

Credit Derivative Options 68

Credit Derivatives Roadmap 116

Credit Derivatives Update 143

 bilateral financial contract which allows specific aspects of credit risk to be

Bond = Duration + Convexity + Credit

Auctioneer sells a number of

Bond / Loan Asset Swap Credit Default Swap

Funding Risk Credit Risk

Risk Free Rate

 Buy Protection  Sell Protection

 “Short Risk”  “Long Risk”

 Pay periodic payments  Receive periodic payments

 Receive contingent payment  Pay contingent payment

 Reference Entity: ABC

 Reference Obligation(s) - The obligation(s)

 Default Requirement: USD 10,000,000 or its

 Obligation Category: Borrowed Money

Reference Entity Risk

X bps per annum

 Buyer decreases exposure to Reference Credit(s), but assumes contingent (“two-

name”) exposure to Seller

$ 100 T + Corporate Spread (SC)

Assumption: Haircut is small ( ≈ 0) & repo rate spread is negligible ( ≈ 0)

Lender AAA Rated Institution

Contingent payment in event of default

Credit Assessment AAA to AA- A+ to A- BBB+ to BBB- BB+ to B- Below B- Unrated

Risk Weight 20% 50% 100% 100% 150% 100%

Credit Derivatives Overview 1

CLN & Linear Basket 14

Credit Derivative Options 68

Credit Derivatives Roadmap 116

Credit Derivatives Update 143

Baskets Portfolio Tranching

Non-capital protected First-To-Default with First Loss (Junior)

 CLNs can be rated and / or listed if required

 Both physical and cash delivery are available to the investor

Interest Rate Swap

Swap Counterparty SPV Collateral

 After the first credit event:

 Advantage of less documentation by taking exposure to many credits in one single

X bp per annum on notional 100

Example: Green Bottle Swap on 5 equally weighted names on a notional of 100

X bp per annum on notional 80

(Par — Recovery on defaulted credit) Protection Seller

= (300 + 100 +400 + 150 +200)/5

Use Credit Default Swaps to maximize liquidity -

Cost efficient and timely access to the Credit Markets

Daily reports on actual versus theoretical pricing

Two major CDS indices trade actively at tight bid-ask spreads

 DJ CDX in North America has 125 names

 DJ iTraxx Europe has 125 names

Can also trade loss tranches on index

 Tranches are like synthetic CDO tranches

 One-factor Gaussian copula is standard for quoting correlations

 iTraxx Australia has 25 names

 CDX.EM has 14 emerging market sovereigns, including Korea, Malaysia and

Credit Derivatives Overview 1

CLN & Linear Basket 14

Credit Derivative Options 68

Credit Derivatives Roadmap 116

bilateral financial contract which allows specific aspects of credit risk to be

Buy Protection Sell Protection

“Short Risk” “Long Risk”

Pay periodic payments Receive periodic payments

Receive contingent payment Pay contingent payment

Reference Entity: ABC

Reference Obligation(s) - The obligation(s)

Default Requirement: USD 10,000,000 or its

Obligation Category: Borrowed Money

Buyer decreases exposure to Reference Credit(s), but assumes contingent (“two-

CLNs can be rated and / or listed if required

Both physical and cash delivery are available to the investor

After the first credit event:

Advantage of less documentation by taking exposure to many credits in one single

DJ CDX in North America has 125 names

DJ iTraxx Europe has 125 names

Tranches are like synthetic CDO tranches

One-factor Gaussian copula is standard for quoting correlations

iTraxx Australia has 25 names

CDX.EM has 14 emerging market sovereigns, including Korea, Malaysia and

Yield enhancement in these structures basically depends on the correlation of the

The first-to-default swap is terminated and Protection Seller has no further

Risk is sourced from the market

First to Default Tranche is sold either in bond or swap form

Bank retains “Senior” Tranche

Each name in the basket behaves

Basket premiums should decline with an increase in correlation. A basket of

Default correlations impact the likelihood of multiple defaults up to a given time

Asset correlations can be extracted from the ability-to-pay process of a portfolio of

given a market environment, the names default independently

As correlation approaches zero: FTD Basket Premium = Sum of Basket Component’s

Investors wishing to maximize premium generated by selling FTD Basket protection

Following a credit event, the dealer will be forced to unwind

The greater this correlation, the greater the expected spread

On the other hand, a low correlation would imply a lower

Basket component correlation premium : FTD baskets that are relatively

First-2-to-default basket swaps allow investors to leverage their exposure to a

The protection buyer is protected against the first two defaults

Equity owner is in the first loss position

Subordinated tranches (including Equity) are leveraged exposures

Senior tranches (AAA, AAAA) are de-leveraged exposures

At the portfolio level, spread compensates for expected loss:

Tranching reallocates losses within the portfolio

The spread on an individual tranche must compensate for the tranche

Default correlation between names — derived from historical asset correlations

We model the correlation between names as a result of the correlation of each

Tranche exposure can be offered in local currency whereas risk is sourced