The Chinese University of Hong Kong

Department of Economics

Econ 5490 Corporate Finance

Topic 2: Bond Prices, Yields, Interest Rates and Term Structure

---------------------------------------------------------------

I. Introduction

One of the main securities that we are discussing in this course is fixed income securities,
that is, debt obligations issued by corporations, e.g., debentures, notes, loans. The value of fixed
income securities primarily depends on: (1) interest rates; and (2) the chance of default by the
issuer. In this course, we will need to know the basics of interest rate risk. Therefore, we will
start, in this note, by thinking about how interest rates affect bonds that are otherwise riskless. A
“riskless” bond is an obligation of the government. Since the government is not very likely to
default, the only relevant risk is interest rate risk. Later in the course we will consider credit risk
or default risk.

Most of the material summarized here can be found in some textbook on capital markets,
for example, Capital Markets: Institutions and Instruments, by Frank Fabozzi and Franco
Modigliani, and Investments, by Zvi Bodie, Alex Kane and Alan J. Marcus.

II. Bond Prices and Yields

A) Simple Bond Pricing

Bond is a major fixed-income security that makes fixed payment, either in nominal or
real terms. We know that to evaluate a security, we discount its expected cash flows by the
appropriate discount rate. The cash flows from a bond consist of coupon payments until the
maturity date plus the final payment of par value. Therefore,

Bond value = Present value of coupons + Present value of par value

If we let C denote the coupon payment, F the par value or principal payment, the interest
rate that is used to discount the future cash flow as r, we have

n
C F
P  (1)
t 1 1  r 
t
1  r n

1
 Each coupon is discounted based on the time until it will be paid. The first term on the
right-hand side is the present value of an annuity. The second term is the present value of a single
amount, the final payment of the bond’s par value.

The present value of a $1 annuity that lasts for T periods when the interest rate equals r is
1 1 1
(1  ). This is called the T-period annuity factor. Similarly, we call the PV
r 1  r n
1  r n
factor.
Thus,
1 1 1
Bond Price = Coupon * (1  )+ Par Value *
r 1  r n
1  r n
= Coupon * Annuity factor (r, T) + Par Value * PV factor(r, T)

B) Bond Yields

The yield to maturity is defined as the interest rate that makes the present value of the cash
flow (coupon payments) equal to the market value (price) of the instrument (bond). If we let the
bond’s yield to maturity be y. Given the specification of cash flows, the relationship between the
bond price and the yield to maturity is given by:

n
C F
P  (2)
t 1 1  y t
1  y n
It is clear that given the coupons, face value, and bond price, the yield is just the IRR
(internal rate of return) that makes the above hold as an equality.

Bonds traded in the U.S. typically make semiannual coupon payments, i.e., every six months.
In this case, we have

n C
F
P 2
t
 n
, (3)
t 1  y   y 
1  2  1  2 
   

where n is now the number of semiannual periods. If one were to solve for the yield of a bond
that pays coupon semiannually, the result would be y/2 that is not directly comparable to annual
yield calculated above because of the difference in compounding periods.

Notice that the yield is calculated by observing the price, P, in the market and solving for y
given the contractual specification of C and F. (Since (3) is a nonlinear function a numerical
procedure is required to get y.) The procedure for calculating y assumes that the promised

2
payments of coupons (C) and principal amount (F) will, in fact, be made, i.e., that there is no
default. In fact, the market price is based on a calculation of the form:

pC  1  P C pF  (1  p) F
P  ,
1  y 
t
(1  y ) n

where p is the probability of default and  is the fraction of the promised amount recovered. In
fact, as we will see later in the course, things are considerably more complicated, but the main
point for the moment is this: a bond yield is a kind of measure that is of little value except in very
special circumstances. One such circumstance would be the case of comparing the return on two
bonds with the same bond rating. But even here there are subtleties making the exercise
dangerous.

C) Coupon Stripping and Arbitrage

Unlike a bond paying coupons, a zero-coupon bond makes only a single payment at
maturity. (These are also called “pure discount” bond.) For the most part, pricing these bonds
presents no problems. Equation (1) or (2) can be used by setting C=0, so that the price is simply the
present value of F. Zero-coupon bonds are either pure discount bonds upon issue or a Treasury
“Strip”— a Treasury bond that has been unbundled and sold in pieces with each piece being a
coupon payment or the principal payment. Unbundling coupon bonds are called “coupon-
stripping.”

Coupon stripping suggests that any coupon-bearing bond can be thought of as a portfolio of
zero-coupon bonds, with each separate payment on the whole bond representing a zero-coupon
bond. This suggests that a sequence of whole bonds (i.e., coupon-bearing bonds) of increasingly
longer maturity can be used to determine zero-coupon prices, even if we do not actually strip bonds.
In fact, for any bond with price P, maturity n, and coupon rate C/F, the price can be represented by:

n
C 
P     z t  Fz n ,
t 1  2 

where zt is the current price of a zero-coupon security with a $1 face value and a maturity of t
periods from now. Clearly, if we have a one-period bond, with any coupon rate, we can solve for z1
if we know the price of a two period bond. And so on.

The relationship between U.S. Treasury bonds and Treasury strips suggests the presence of
an arbitrage relationship. Since the whole bond can be reconstituted from strips, the whole bond
and the replicating portfolio of strips should sell at the same price. If they do not, then there is an
arbitrage opportunity.

3
Example: A Treasury note with an 8 1/8% coupon that matures in February 1998. Coupons are
payable on February 15 and August 15 of each year. On June 9, 1994 the asking price of the bond
was 105:29, i.e., 105 29/32 or 105.90625 per $100 face value. U.S. Treasury bonds are quoted on a
“skip-day settlement” basis. This means that a trade occurring on June 9 would be settled on the
second business day after the trade, or on June 13, June 13 is 118 days from the last coupon date,
while the entire coupon period, from February 15 to August 15, is 181 days. Thus, in addition to the
quoted price, the bond buyer must pay the seller 118/181 of a coupon payment. Therefore, the
invoice price of the bond is $105.90625+(118/181)(4.0625)=108.55473.

To see if the 8 1/8% notes of February 1998 afford an arbitrage opportunity, we can use
Treasury strip prices from June 9 to see if we could sell the pieces for more than the whole bond.
Since we are selling these pieces, we should use “bid” instead of “ask” prices.

Date of Payment Amount Bid Price Per $1 of Proceeds

Payment
Aug. 1994 4.0625 0.993125 4.03457
Feb. 1995 4.0625 0.9690625 3.93682
Aug. 1995 4.0625 0.9421875 3.82764
Feb. 1996 4.0625 0.910625 3.69941
Aug. 1996 4.0625 0.8815625 3.58135
Feb. 1997 4.0625 0.8515625 3.45947
Aug. 1997 4.0625 0.821875 3.33887
Feb. 1998 4.0625 0.79375(coupon) 3.22461
Feb. 1998 100.00 0.7940625(Principal) 79.40625
Total 108.50898
Whole Bond Ask 108.55473
Price
Arbitrage Profit -0.04575

These calculations show that the prices of the notes and strips are very close. In fact, a trader who
bought the notes and stripped them would actually lose about a nickel per $100 face value.

III. Duration

One of the risks facing a bondholder is the chance that interest rates change. A change in
interest rates has two effects. On the one hand, the value of the bond moves inversely with interest
rates (“price risk”). On the other hand, the value of reinvesting the coupons moves positively with
interest rates (‘reinvestment risk”). In other words, if rates rise, the value of a bond declines, but the
opportunity to reinvest the coupons as they are paid becomes more valuable. One definition of
duration is that it is the horizon over which these two risks would exactly cancel.

4
 Duration is a measure of the sensitivity of a bond’s price to a change in yield. It is derived
by differentiating the equation given above for the price of a bond, (2) or (3). This gives
MacCaulay Duration, DMAC, as:

n
tC nF

dP
 1  y   1  y 
t 1
t n
P D
MAC  (4)
d (1  y ) P
(1  y )

MacCaulay duration is the percentage change in the price of the bond with respect to the percentage
change in 1 plus the yield. Another way to express this is to say that duration is the elasticity of the
bond’s price with respect to one plus the yields. Note that duration is a weighted average of all the
time to payment, with weights equal to the present value of the payment at that date divided by the
bond price. In other words, duration is a weighted-average term to maturity of the components of a
bond’s cash flows (coupon payments and principal payment), in which the time of receipt of each
payment is weighted by the present value of that payment. The denominator is the sum of the
weights, which is precisely the price of the bond.

An alternative form of MacCaulay Duration is:

C 
1  y  n  y 
1 y F 
DMAC   (5)
y C
F
 
1  y n  1  y

A closely related concept is Modified Duration, DMOD, which is defined as:

 dP
D
DMOD  P  MAC (6)
dy 1  y 

Modified Duration has the natural interpretation that the percentage change in a bond’s price when
its yield changes by dy is given by:
dP
  DMOD dy (7)
P

Sometimes this is expressed as dP/dy= -PDMOD.

So, we have

5
Percentage price change = - Modified duration * Yield change

Example: Consider an 8% bond with yield to maturity of 10% (5% each half year) and a maturity
of two years. We calculate duration of this bond.

(1) (2) (3) (4) (5)

Time Until Payment Payment PV of (3) divided (1) times (4)
Payment (yrs) Discounted at by bond price
5%
semiannually
0.5 $ 40 $38.095 0.0395 0.0198
1.0 $ 40 36.281 0.0376 0.0376
1.5 $ 40 34.553 0.0358 0.0537
2.0 $ 1040 855.611 0.8871 1.7742
SUM $964.540 1.00 1.8853

The bond has a maturity of two years but a duration of 1.8853 years.

You should convince yourself that:

1) The duration of a zero-coupon bond equals its maturity.

2) Holding maturity constant, a bond’s duration is higher when the coupon rate is lower.
3) Holding the coupon constant, a bond’s duration generally increases with its time to maturity.
Duration always increases with maturity for bonds selling at par or at a premium to par.
4) Holding other factors constant, the duration of a coupon bond is higher when the bond’s
yield is lower.

IV. The Term Structure of Interest Rates

The yield curve or term structure of interest rates depicts the relationship between yields on
instruments of different maturity. There is no unique relationship between yield and maturity for
coupon-bearing bonds, whereas there is a unique relationship for zeros. In addition, as we saw
above, the only securities whose duration is equal to their maturity are zero-coupon bonds.
Therefore, the first step in constructing the term structure then is to construct zero-coupon yields for
various maturities. This can be done by first deriving implicit zero prices from coupon bonds as we
did above. From these we calculate the yields, y1, y2,…., yn, for zero-coupon bonds with maturities
of 1,2,…., n periods. These yields are referred to as “spot” rates. The relationship between time to
maturity and the yields for zero-coupon bonds of different maturity is called the “spot rate curve,”
“yield curve,” or “spot yield curve”.

A) Forward Rates

6
 The future interest rate is typically uncertain. But we may think that market participants can
form a market’s consensus for future interest rate.

Consider the following two investment alternatives for an investor who has a one-year
investment horizon:

Alternative 1: investor buys a one-year instrument

Alternative 2: investor buys a six-month instrument and when it matures in six months buys
another six-month instrument.

With Alternative 1, the investor will realize the one-year spot rate and that rate is known
with certainty. In contrast, with Alternative 2, the investor will realize the six-month spot rate, but
the six-month rate six months from now is unknown. Therefore, for Alternative 2, the rate that will
be earned over one year is not known with certainty.

The investor will be indifferent to the two alternatives if they produce the same total dollars
over the one-year investment horizon. Given the one-year spot rate, there is some rate on a six-
month instrument six months from now that will make the investor indifferent between the two
alternatives. We will denote that rate as f.

Suppose the six-month spot rate is y1 and the one-year spot rate is y2. We define f as the six-
month interest rate that prevails six months from now.

The proceeds from investing $100 through Alternative 1 is

$100*(1+y2) 2
Remember we are working in six-month periods, so that subscript 2 for y2 represents two six-
month periods, or one year. When buying a one-year bond, investor has set the interest rate for the
two six-month periods at the spot rate (current interest rate) y2 already.

Total dollars at the end of six months for Alternative 2 is

$100*(1+y1)
where y1 is the six-month spot rate. If this amount is reinvested at the six-month rate six months
from now at a rate f, then the total dollars at the end of one year would be:
$100(1+y1)(1+f)

The investor will be indifferent between the two alternatives if the total dollars are the same,
that is,
$100(1+ y2) 2 = $100(1+y1)(1+f)

Solving it, we get

(1  y 2 ) 2
f2 = 1
(1  y1 )

7
More generally, the forward rate for a one-period bond to be bought n periods from now is defined
by:

(1  y n ) n
fn = 1
(1  y n 1 ) n 1

Similarly, forward rates for bonds of more than one period are defined analogously. For example,
given the spot rates y7 and y4 for seven and four-period bonds, respectively, we can define the
forward rate, 3f7, for a three-period loan that matures at time 7:

1  y7 7  1  y 4 4 1  3 f7  .
3

B) Theories of the Term Structure of Interest Rates

There are several theories that have been put forward to explain the shape of the term
structure and how it evolves through time. However, these are omitted for this course.

V. Points to Remember:

 Bond price and yield

 Check for arbitrage between strips and whole bonds
 Calculate zero-coupon bond prices and spot yield curve
 Calculate Duration
 Understand properties of Duration
 Forward rates