12 Overview of Default Risky Bonds

12.1 Background

In most of our discussion topics in fixed income securities we have as-

sumed that the parties involved in the contractual obligations of the secu-
rity will honor their obligations. The main risk present in these securities is
the stochastic nature of the interest rate. One of the parties to the contract
may, however, default on its obligations. For example, an issuer of a cor-
porate bond may be unable to make the required payments of coupons
and/or the principal payments. The contractual obligation in a financial se-
curity may not be a simple bond like structure and when a party to such a
contract can default the pricing problem becomes complex. For example,
the bond issuer’s credit rating might be downgraded and there may not be
any comparable security to look up for a price reference.
A formal model to price such defaultable security based on their ex-
pected payoffs is thus a necessity. Since a typical financial institution may
hold in its portfolio a range of defaultable securities it is important for the
institution to be able to measure the aggregate credit exposure in its portfo-
lio. This may require reliably estimating possible values of such securities
at various times in the future. The management of such credit risks is also
a requirement imposed by the central banks so that the financial institution
can allocate sufficient capital for the prudent management.
We discuss in this topic some of relevant aspects of the defaultable se-
curities and in particular the bonds. We need to understand how a company
can ascertain the probability that a counter party will default. In this con-
text the concept of risk-neutral and real world probabilities become impor-
tant to understand. The next important aspect we need to understand how
to estimate the loss given that the default occurs. Thus, the two quantities,
the probability of default and the loss given default jointly determine the
financial institution’s expected loss.
2 12 Overview of Default Risky Bonds

12.2 Credit Ratings in Practice

Moody’s and S&P are the most widely known rating agencies in the world.
There are possibly smaller organizations in some areas. The bond issuers
are allocated different rating classes depending on the perceived ability of
the issuer to meet future payment obligations. For example, Moody’s allo-
cate a rating of Aaa to denote the best corporate rating implying almost no
probability of default. The next best rating is given by Aa, and then comes
A, Baa, Ba, B, and Caa. In practice only bonds with ratings Baa or above
are considered to be of investment grade. The S&P has also similar rating
classes using different naming convention. Moody’s increases the granu-
larity of the rating classes by introducing addition classes such as Aa1,
Aa2 and Aa3 within the rating class Aa. Similarly, there are classes like
A1, A2 and A3 within the A class bonds. Further details of such classifica-
tion may be found on the respective organizations’ websites.
Bond traders take into account of the rating class of a particular issuer to
come up with a price for a bond issued by such an issuer. The construction
of zero coupon yield curve for Treasury securities is well established.
Similar approach may be adopted to construct a basic zero coupon yield
curve for a given rating class from the market prices of actively traded
bonds of this rating class. This basic zero coupon yield curve for the par-
ticular rating class is then used to price any newly issued bond of the same
class. There is some theoretical issue with this assumption and the inter-
ested reader may consult Hull (2003), chapter 26.
In figure 12.1 the typical appearance of the zero coupon yield curves for
different rating classes over the corresponding Treasury zero-coupon yield
curve is shown. It is quite obvious that the spread expands as the credit rat-
ing goes down. Besides, as the maturity of the bond increases the spread
for lower rated bonds expand at a faster rate.

12.3 Losses Expected Due to Default

Here we discuss how we are able to quantify the default loss from the
available bond prices under some simplifying assumptions. For a given
maturity we can compare the prices of Treasury issued zero-coupon bond
with the prices of similar bonds issued by corporations of different rating
classes. The lower prices of corporate bonds may be thought of as due to
the effect of losses inherent in such bonds. We then express this price dif-
ference as a percentage of the Treasury (or no default) prices. Although
 12.3 Losses Expected Due to Default 3

there are some theoretical issues in the assumption that the price difference
is entirely due to credit risk, we will continue with this approach.

2.5

1.5

0.5

0
3m 6m 2y 7y 10y

AAA AA BBB1 B1

Fig. 12.1 Yield spread for different rating classes as a function of maturity

In essence we compute the price of a Treasury zero coupon bond using

the Treasury yield curve and similarly use the corporate yield curve to
price the corporate zero coupon bond. Hull (2003, page 612) indicates why
sometime practitioners use LIBOR yield curve to price the no default zero
coupon bonds. In any case, the percentage price difference becomes the
measure of the expected loss due to default from corporate bonds.
In the Tables 12.1 and 12.2 below we show this computation for two
different rating classes and for five different maturities. This has been done
for two different dates, March 1997 and March 2001. From the computed
expected loss we may think of that both corporate rating classes did be-
come more risky in 2001 compared to March 1997. These could be due to
several economic as well as historical reasons. We do not, however, plan
to pursue that in this discussion.
4 12 Overview of Default Risky Bonds

Table 12.1 Default Loss Expected (March 1997)

Maturity Treasury AAA BBB1 Default Loss Expected

Year Yield % Yield % Yield % % of No Default Value
AAA BBB1
0.25 5.35 5.52 5.81 0.042 0.115
0.50 5.55 5.80 6.04 0.125 0.245
2.00 6.45 6.61 6.78 0.319 0.658
7.00 6.89 7.14 7.37 1.735 3.304
10.00 6.92 7.16 7.50 2.371 5.635
All yields are assumed continuously compounded.

Table 12.2 Default Loss Expected (March 2001)

Maturity Treasury AAA BBB1 Default Loss Expected

Year Yield % Yield % Yield % % of No Default Value
AAA BBB1
0.25 4.30 4.73 5.57 0.107 0.317
0.50 4.09 4.58 5.40 0.245 0.653
2.00 4.18 4.83 5.69 1.292 2.975
7.00 4.86 5.61 6.57 5.115 11.281
10.00 4.93 5.83 6.83 8.607 17.304
All yield are assumed continuously compounded.

12.4 Default Probability with No Recovery

The default probability is an important parameter in dealing with default

risky bonds. Since default events are rare, there is not much historical data
available on default events. Thus, the estimation of default probability
based on historical data turns out to be imprecise. This section explores the
information content in the yields on bonds issued by corporations. This
approach is typically referred to as the reduced-form model. The main as-
sumption is that an exogenous random variable drives the default event
and that the probability of default over any period is non-zero. This may
vary over time as well. In the reduced-form model the emphasis is placed
on the probability distribution of default. Also, if a corporation does not is-
sue bonds, the bond yield data from another entity of similar risk class may
be used to infer the default probability.
 12.4 Default Probability with No Recovery 5

Let us define the following quantities in order to quantify the default

probability:
y ( T ) : Yield on a corporate zero coupon bond of maturity T
y 0 ( T ) : Yield on a zero coupon Treasury bond of maturity T
q ( T ) : Probability of corporate default between now and T
With the help of these variables we can express the value of a Treasury
zero-coupon bond today as (assuming $1 face vale and continues com-
pounding) as exp ( − y o ( T ) × T ) . Similarly the value of a corporate zero-
coupon bond today is exp ( − y ( T ) × T ) . Since we are assuming zero re-
covery if the corporation defaults then the payoff from the corporate bond
at T would be 0 with probability q ( T ) and the face value $1 with prob-
ability (1 − q ( T ) ) . Therefore, the value of the corporate bond today
would be the expected payoff at T discounted by the risk-free zero-coupon
yield. This discounting by the risk-free zero-coupon yield implies that the
probabilities are risk-neutral probabilities. Since the yield on the corporate
bond is y ( T ) , we can write,

exp ( − y ( T ) × T ) = ⎡⎣q ( T ) × 0 + (1 − q ( T ) ) × 1⎤⎦ exp ( − y 0 ( T ) × T ) (12.1)

q ( T ) = 1 − exp ⎡⎣ − { y ( T ) − y 0 ( T )} × T ⎤⎦ . (12.2)

With the help of the equation (12.2) we can now compute the cumula-
tive default probability for the data shown in Tables 122.1 and 12.2. These
are shown below in Tables 12.3 and 12.4. The last column in these tables
shows the probability of default in the two consecutive years given in the
first column. This is computed as the difference between the two consecu-
tive cumulative probabilities. We also note that the cumulative probability
is the same as that computed as the default loss expected in Tables 12.1
and 12.2. These results are, of course, under the assumption of no recovery
in case of default.
6 12 Overview of Default Risky Bonds

Table 12.3. Cumulative and Marginal Probabilities of Default (March 1997)

Maturity Yr. Cumulative Probabilities Marginal Probabilities

AAA BBB1 AAA BBB1
0.25 0.042% 0.115% 0.042% 0.115%
0.50 0.125% 0.245% 0.082% 0.130%
2.00 0.319% 0.658% 0.195% 0.413%
7.00 1.735% 3.304% 1.415% 2.646%
10.00 2.371% 5.635% 0.637% 2.331%

Table 12.4. Cumulative and Marginal Probabilities of Default (March 2001)

Maturity Yr. Cumulative Probabilities Marginal Probabilities

AAA BBB1 AAA BBB1
0.25 0.107% 0.317% 0.107% 0.317%
0.50 0.245% 0.653% 0.137% 0.336%
2.00 1.292% 2.975% 1.047% 2.322%
7.00 5.115% 11.281% 3.823% 8.306%
10.00 8.607% 17.304% 3.492% 6.023%

12.5 Alternative Expressions of Default Probabilities

While maintaining the assumption of no recovery, the notion behind the

default probability is clear from the forgoing discussions. In this section
we outline the two different but related definitions normally used in prac-
tice to express this quantity.
The two definitions are termed a) hazard rates and b) default probability
density. The term hazard rate is borrowed from the insurance industry and
denotes the probability of default over a small interval of time in future
provided no default occurred before. If h ( t ) denotes the hazard rate at
time t then the probability of default between time t and t + δt is given by
h ( t ) δt provided no default occurred until time t. Similarly, the default
probability density, q ′ ( t ) , defines the unconditional default probability
between t and t + δt as q′ ( t ) δt , inferred at time zero. These two quanti-
ties are elated by the following,
 12.6 Introducing Recovery Rates 7

( )
q′ ( t ) = h ( t ) exp − ∫ h ( τ ) dτ .
t

0
(12.3)

Thus, either the hazard rate or the default probability density could be
used in quantifying this phenomenon. We can illustrate this concept using
the data in Table 12.4. The marginal probability of default for the AAA
bond issuer in year 10 is 3.492%. The probability of no default until year 7
is (1-0.05115) = 0.94885. Thus the hazard rate for default between year 7
and year 10 is (0.03492/0.94885) = 0.03680.

12.6 Introducing Recovery Rates

In the discussions so far we have assumed that when the company defaults
the amount recovered is zero. In practice though there is always some
amount is recovered by the lenders depending upon the seniority of the se-
curity held by a particular investor. We are not, however, involved in ana-
lyzing the time it might take through the legal system to recover any such
amounts.
We simply focus on the concept of the recovery rate of the claimable
amount of the debt. For the bond like security it is possible to assume that
the claimable amount is a percentage (R) of the no-default value of the
bond. Thus the recovery rate is assumed to be R.
Keeping our focus on the zero coupon bonds and under the assumption
that the default can only occur at the maturity date we can state that if there
is no default the holder receives $1 (face value of the bond) and the value
of this today is exp ( − y ( T ) × T ) . If the bond defaults with a probabil-
ity q ( T ) , the received amount is R × e xp ( − y 0 ( T ) × T ) . Thus, we can
write,

exp ( − y ( T ) × T ) = ⎡⎣ q ( T ) × R + (1 − q ( T ) ) × 1⎤⎦ exp ( − y 0 ( T ) × T (12.4)

This simplifies to give,

8 12 Overview of Default Risky Bonds

q (T) =
( )
1 − exp − ⎡⎣ y ( T ) − y 0 ( T ) ⎤⎦ T
.
(12.5)

1− R

It should be noted that this expression applies under the assumption of

fixed recovery rate and the claimable amount is the no default value of the
bond. Besides, the default is assumed to occur only at the maturity of the
bond. In the next section we will relax some these assumptions and attempt
to make the model more realistic.

12.7 Multiple Bonds Issued and Realistic Scenario

The discussion in this section follows the ideas in Hull (2003) and the
spreadsheet accompanying this topic shows how the models are imple-
mented on Excel. The earlier analysis was based on the zero coupon bonds
issued by a company. Since such corporate bonds are almost non-existent
we discuss in this section corporate coupon paying bonds of different ma-
turities. We will also make two different assumptions about the claimable
amount if default occurs. We, however, maintain the assumption that a de-
fault can only occur at a bond maturity.
In order to quantify the results of this analysis and also to help the
reader implement the model on a spreadsheet, we need several symbols
with specific meanings to be defined. These are itemized below.

− BJ : Today’s price of the Jth bond issued by the firm.

− G J : Today’s price of a default-free bond with cash flow characteristics
same as that of the Jth corporate bond.
− FJ,I : Forward price of the G J bond at the Ith time
− d I : Present value of $1 paid at the Ith time at the risk-free rate
− CJ,I : Claimable amount in case of default of the Jth bond at the Ith time
− R̂ :Recovery rate applied to the claimable amount in case of default
− a J,I : Present value of the default loss for the Jth bond at the Ith time
− pI : Probability of default at the Ith time

Certain other technical assumptions are needed for the analysis below to
hold and these are pointed out in Hull (2003, page 615).From the defini-
 12.7 Multiple Bonds Issued and Realistic Scenario 9

tions of the terms above it is clear that the present value of the loss if there
is a default at time I, then,

a J,I = d I ⎡⎣ FJ,I − RC
ˆ ⎤
J,I ⎦ .
(12.6)

Since pI is the probability of default at time I, the total present value of the
losses from the Jth corporate bond may be written as,

J (12.7)
G J − BJ = ∑ pi a J ,i .
i =1

The recursive nature of this equation allows us to compute the probabilities

sequentially starting from the very first one as,

G1 − B1 (12.8)
p1 = .
a1,1

The remainder of the probabilities could be obtained from the equation be-
low:

J −1 (12.9)
G J − BJ − ∑ pi a J ,i
pJ = i =1
.
a J,J

The recursive nature of this equation makes it easily implementable in a

spreadsheet. The following sample of corporate bonds has been analyzed
following the foregoing method and the results tabulated below. The de-
tails of the computation are in the associated spreadsheet.
The sample used for illustration assumes a 30% recovery rate of the
claimable value. The two different claimable amounts are considered i.e.
the no-default value of the bond and the ace value plus the accrued inter-
est. For the sake of simplicity we have also assumed the risk-free rate to be
constant over the sample period of analysis. The methodology also needs
computing the value of forward bond price and this is briefly outlined be-
10 12 Overview of Default Risky Bonds

low. Additional details about these contracts could be found in Hull

(2003). Assume that the current price of a security is S and we are inter-
ested in estimating the forward price of this security at time T in future.
The security also provides cash income during this period. The present
value (computed using the risk-free rate) of all these cash income from the
security is I, then the forward price (F) is given by,

2×T (12.10)
⎛ r⎞
F = (S − I ) ⎜1 + ⎟ .
⎝ 2⎠

The above relation assumes that the risk-free interest rate is r p.a. and it is
compounded half-yearly basis. This is a common assumption since the
corporate bonds pay half-yearly coupons.
The sample of corporate bonds analyzed is described in the table below:

Table 12.5. Details of Corporate Bonds

Bond Maturity (Years) Coupon (% p.a.) Bond Yield (% p.a.)

1 7.00 6.60
2 7.00 6.70
3 7.00 6.80
4 7.00 6.90
5 7.00 7.00
7 7.00 7.20

The analysis assumes that the risk-free rate is at 4.75% p.a. through out the
sample period. The computed probability of default is given in the follow-
ing table for two different assumptions about the claimable amount in case
of default and 30% recovery rate.

Table 12.6. Default Probability

Bond Maturity (Years) No-Default Value Face Value Plus

Accrued Interest
1 0.0259 0.0259
2 0.0282 0.0280
3 0.0305 0.0300
4 0.0326 0.0320
5 0.0347 0.0338
7 0.0814 0.0755
 12.10 Risk Neutral Valuation 11

12.8 Defaults at Any Time

In this section we just outline the modification needed in the analysis given
in the previous section when we allow the bonds to default at any time and
not just on bond maturity. It is intuitively clear that the problem is much
more complex in this situation. Additional details could be found in Hull
(2003) as well as in Schönbucher (2003).
In this situation we have to deal with the default probability density
q ( t ) as discussed earlier. If we assume that this is constant over the inter-
′
val t I−1 < t < t I , then the equivalent expression for (12.6) is,

tI (12.11)
b J,I = ∫ d t ⎡⎣ FJ,t − RC
ˆ ⎤
J,t ⎦ dt .
t I−1

The expression equivalent to (12.9) would then become.

J −1 (12.12)
G J − BJ − ∑ q′i b J,i
q′J = i =1
.
b J,J

The integral in equation (12.11) could be evaluated using Simpson’s rule

as pointed out in Hull (2003).

12.10 Risk Neutral Valuation

The foregoing analysis of estimating default probabilities is referred to be

risk-neutral probabilities. As Hull (2003, page 620) describes that the es-
timated probabilities from bond prices turn out to be much higher than that
have been observed in the historical data on corporate defaults. The differ-
ence lies in the fact that we are dealing with two different probability
measures. Thus the question naturally arises about which probability
measure to use? It actually depends upon the use of the computed prob-
abilities. Since all derivatives valuation takes place in the risk neutral
world, we need to use the risk-neutral default probabilities to price credit
risky derivatives. However, if we are interested in e.g. scenario analysis
then we need to use real world or historical probabilities.
12 12 Overview of Default Risky Bonds

As this concept of risk neutral valuation is at the heart of derivatives

pricing models, we present in this section a basic understanding of this im-
portant principle. Let us focus on an economy where the uncertainties are
represented by two different outcomes in one period of time. This is same
as the binomial modeling approach used in option pricing literature. We
further assume that there are only two different securities, one risk-free
and the other risky security. The risk-free security pays $1 irrespective
which state occurs next period and its current value is b 0 . On the other
hand the risky security pays v1 in state 1 next period and v 2 in state 2 next
period. Its current value is v 0 . These two payoffs will in general be differ-
ent since investors would demand different returns depending on the per-
ceived risks of the two states that could occur next period.
Now, let us introduce the notion of the elementary prices. These are also
referred to as Arrow-Debreu securities. This is defined as the price of a se-
curity today that pays $1 in a given state next period and $0 in the other
state. This definition could be extended to multiple states occurring next
period. Thus, in this case we have two elementary prices denoted by, p 0,1
i.e. the price of a security today that pays $1 in state 1 next period. Simi-
larly, we can define p0,2 . With these definitions we can write,

b0 = p 0,1 + p0,2
.
v 0 = p0,1 × v1 + p0,2 × v 2

We can now re-express these in slightly different form as,

⎡ p0,1 p0,2 ⎤
b0 = ( p0,1 + p0,2 ) ⎢ + ⎥
⎣⎢ p0,1 + p 0,2 p0,1 + p0,2 ⎦⎥

⎡ p 0,1 p0,2 ⎤
v 0 = ( p0,1 + p0,2 ) ⎢ v1 + v2 ⎥.
⎣⎢ p0,1 + p0,2 p0,1 + p0,2 ⎦⎥

We should note from the definition that the gross risk-free return ( r0 ) is
1 p 0,1 p0,2
given by . We also define, q1 = , q2 = .
p0,1 + p0,2 p0,1 + p0,2 p0,1 + p0,2
Then the above expressions simplify to,
 12.11 Merton’s Model of Default Risky Bonds 13

1
b0 = ( q1 + q 2 ) ,
r0
1
v0 = ( q1 × v1 + q 2 × v 2 ) .
r0

The implication of the above expressions is profound once we recognize

that both q1 and q 2 are bounded between 0 and 1 and these two sum up to
1. Thus these two quantities satisfy the requirement to be probabilities of
the two states occurring next period. Furthermore, the current value of the
securities are computed as the expected value of the next period payoffs
but discounted by the risk-free rate. In other words, these two new prob-
abilities have the effect such that the investors are suddenly risk-neutral.
This is the basic insight into the risk-neutral valuation principle applied
to pricing derivative securities. Additional mathematical details may be
found in Hull (2003).

12.11 Merton’s Model of Default Risky Bonds

Merton (1974) used the option pricing framework to develop a model for
pricing the corporate bonds. The model assumes that a firm has a single
zero-coupon bond in its capital structure. The firm has promised the bond
holder to repay the face value of the bond on its maturity date. If the firm’s
value on the bond’s maturity date is less than the face value of the bond,
the firm actually defaults in its promised payment. It also assumes that the
claim of the bond holder is superior to that of the equity holder. Thus, if
the firm’s market value at maturity is greater than the face value of the
bond the debt holder gets the full face value back. On the other hand, if the
firm’s value is less than that of the face value of the bond the debt holder
gets the lower amount. This payoff pattern of the risky debt at maturity re-
sembles that of an option where the underlying asset is the firm’s value
and the exercise price is the face value of the debt.
We can now quantify this idea and explain how Back-Scholes option
pricing model might be utilized to compute the value of the risky bond.
We need the following symbols for this purpose and follows the presenta-
tion in Hull (2003).

− V0 : Value of the firm’s asset today

14 12 Overview of Default Risky Bonds

− VT : Value of the firm’s asset at time T

− E 0 : Value of equity today
− E T : Value of equity at time T
− D :Total amount of debt to be repaid at time T
− σV : Volatility of assets of the firm
− σ E : Volatility of the firm’s equity
Following the argument in the previous paragraphs, Merton’s sugges-
tions lead to the following expression of the equity value of the firm at
time T,

E T = max ( VT − D, 0 ) . (12.13)

The equity is thus a call option on the value of the firm’s assets with the
exercise price of D. Assuming that the Black-Scholes framework applies
here, we can write the value of the firm’s equity today as the price of the
European style call option. This gives,

E 0 = V0 N ( d1 ) − De − rf T N ( d 2 ) , (12.14)

where rf is the risk-free interest rate for the same maturity. N ( i ) repre-
sents the cumulative normal distribution with the appropriate argument.
These arguments are defined as,

1 ⎡ ⎛ V0 ⎞ ⎛ σ2V ⎞ ⎤ (12.15)
d1 = ⎢ln ⎜ ⎟ + ⎜ rf + ⎟ T⎥
σV T⎣ ⎝D⎠ ⎝ 2 ⎠ ⎦.
d 2 = d1 − σ V T

The value of the debt today is, therefore, V0 − E 0 .

The interpretation of the second term in the equation (12.14) is that it is
the present value of the exercise price conditional upon the option is in the
money or when VT > D . It can also be demonstrated that the probability of
the option being in the money is given by N ( d 2 ) . Alternatively, the prob-
 12.12 Conclusions 15

ability that the company will default is 1 − N ( d 2 ) ≡ N ( −d 2 ) . As we out-

lined before, this is the risk-neutral probability rather than the real world
probability. To make this model operational we need good estimates of
V0 and σV . However, these quantities are not easily observable in the
market. Hull (2003) suggests the following mechanism to get around the
problem. Under certain assumptions and using Ito’s lemma it could be
shown that,

σ E E 0 = N ( d1 ) σ V V0 . (12.16)

For publicly traded firms the equity value and its volatility are quite easily
measurable from the market observations. Thus using the equations
(12.14) and (12.16) V0 and σV may be jointly estimated by solving these
two non-linear equations. This essentially becomes a non-linear optimiza-
tion problem under a suitable norm. The example in Hull (2003) clarifies
this issue further.

12.12 Conclusions

The valuation model of defaultable securities started by Merton has been

improved by various researchers. The research in this area and credit de-
rivatives are going on and continuous improvements are regularly re-
ported. One of the important sources of information in this field is the
website at WWW.DefaultRisk.Com. This is a great resource for any one
interested in this field. However, before concluding this topic here we
point out the main shortcomings of the Merton’s model.
First, the Merton’s model assumes the firm only defaults at maturity of
the debt. This is clearly not a very realistic assumption. Second, in order
for the model to handle various classes of debts issued by a firm the sen-
iority/priority structures of the debts have to be clearly specified. The
model assumes that such priority rules are strictly adhered to in case of de-
faults. There are, however, empirical evidences that this does not always
happen. We have already mentioned and discussed some practical solution
to the problem of non-observabitity of the two important inputs namely the
current market value of the firm and its volatility. This problem becomes
more serious when the firm has issued different classes of debts. Also, the
16 12 Overview of Default Risky Bonds

market value of a firm may contain intangible components such as brand

names etc.
Such shortcomings of the model have led to the development of other
approaches. The references cited below are a good source of information
of these different approaches.

12.13 Examples and Exercises

Exercise 1:
Suppose that a company has issued four bonds of different maturities
and coupon rates as tabulated below. All coupons are paid half yearly. The
yield spread over the risk-free rate of the four bonds is also given in the ta-
ble. The risk-free rate is assumed to be flat at 6% p.a. Assume that today’s
date is 13 July 2004. Estimate the probability of default for these bonds
under the assumptions that the bonds can only default on the maturity
dates and the claimable amounts are face value plus the accrued interests.

Maturity Yield Coupon Recovery

Dates Spread Rate Rate
13-Jul-06 2.00% 9.00% 0.25
13-Jul-08 2.30% 10.00% 0.25
13-Jul-10 2.50% 10.00% 0.25
13-Jul-15 3.00% 12.00% 0.25

References

Merton, R. (1974), On the pricing of Corporate Debt: The Risk Structure of Inter-
est Rates, Journal of Finance, 29, pp. 449-470.
Bluhm, C., Overbeck, L. and Wagner, C. (2002), An Introduction to Credit Risk
Modeling, Chapman & Hall.
Hull, J. (2003), Options, Futures and Other Derivatives, Fifth edition, Prentice
Hall.
Schönbucher, P. J. (2003), Credit Derivatives Pricing Models, Models, Pricing and
Implementation, Wiley Science.