Yield Curve

t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
h, on) orat 201 anc o W Ed Cor ), Ju te F 17. e (4 elc itio rpo
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
(4 elch itio por ly 2 inan Ivo th E h, C n), rate 017 nce
4t
,
Iv
th
Fi
C
y
d
o
n
01
ul
W
v
4
rp
h,
te
o
7.
a
I
(
,J
th
2
n
an vo
a
t
o
01
.
di
n
i
lch ion or
C
y
F
o
.I
na
ul
W
E
y2
Iv
p
,
e
o
h
a
ce
J
r
at
th
i
y2
in
o
c
F
t
o
.
),
i
l
Iv
r
C
d
e
20
F
e
o
l
W
n
E
t
Ju
p
,
n
a
7.
ch
a
or
th
ly
Fi
2
or
in
o
Time-Varying Rates of Return and the Yield Curve
c
t
01
(4 elch itio por ly 2 nan Ivo th E h, C n),

di
l
n
Iv
Ju
,C
ly
4
e
F
4t elch tion pora y 20 nan Ivo th E , Co ), J ate
o
na
W
(
oW E
y2
Ju
p
io
,
ce
or
)
h
i
2
r
o
F
t
n
tio po
ul
4t
di
n
Iv
po ly
lch io
e
0
e
o
When Rates of Return are Different
a
e(
E
2
or
at
Ju
p
in
o
.
In this chapter, we will make the world a little more complex and a lot more realistic,
7
r
h
ly
i
or
C
c
c
t
o
although we are still assuming perfect foresight and perfect markets. The first
n
01
i
l
n
v
Ju
r
4
d
e
rp
assumption that we will abandon is that rates of return are the same no matter what
,
I
na
W
e(
2
the investment time horizon is. In the previous chapters, the interest rate was the
o
a
at 201 nce o W Ed Cor , Ju e Fi 17.

,
i
same every period—if a 30-year bond offered an interest rate of 5% per annum, so
r
)
ly
i
Ed
iti
r
an vo
c
c
F
o
n
po
did a 1-year bond. But this is not the case in the real world. Rates of return usually
l
an
(4
d
lch io
e
20
te
vary with the length of time an investment requires. I
n
h
W
h
or
t
n
elc
(4 elc itio
a
The U.S. Treasury Department Resource Center informs you of this fact every day.
.
t
di
ce
20 anc o W h E , Co ), J te F 17
(4
ly
r
C
For example, on Dec 31, 2015, a U.S. Treasury paid 0.65% per annum for a payment
c
n
7. e (4 elc itio po
W
l
u
to be delivered in one year (Dec 31, 2016), but 3.01% per annum for a payment to
d
o
e
0
,
W
(
be delivered in 30 years (Dec 31, 2045):
elc iti
2
t
n
vo
h
),
ce
7
4
i
Ed
r
e
01
tio po
l
Fi 7. I
(
W
In Percent Per Annum
4
,
e
1m 3m 6m 1y 2y 3y 5y 7y 10y 20y 30y
W
vo th
(
2
r
nc
a
o
7.
ce
12/31/2015 0.14 0.16 0.49 0.65 1.06 1.31 1.76 2.09 2.27 2.67 3.01
)
(4
i
r
,C
F
a
01
o
t
W
u
n
Is this stuff that just bond traders need to know? Not at all. In fact, this stuff matters
e
7.
h
h
to you, too. Have you ever wondered why your bank’s one-month CD offered only
c
.
t
1
e
7
0.21% per annum, while their five-year CD offered 0.8% per annum? (These were
4
or
0
c
F
on
01
I
(
the average national rates on Dec 31, 2015.) Which should you choose? Why? And
u
y2
p
h,
an Ivo th E , C n), J ate

.
01 nce
does inflation matter?

J
h
i
7
elc iti
r
F
elc
7.
CEOs must also know how to compare what they should have to pay for investors
t
1
,
l
)
v
Ju
willing to give them money for a return promise in a 1-year project vs. a 30-year
F
y 2 na
1
I
W
t
project. If the U.S. Treasury has to offer higher rates for lengthier investment return
lch io
0
ra
o W th E , Co ), J ate
7.
h
h
i
or
periods, surely so will firms!

F
t
an vo
o
di
l
In this chapter, you will learn how to work with time-dependent rates of return and
4
or , Ju e Fi
u
p
,C
e
I
(
inflation. In addition, this chapter contains an (optional) section that explains the
o W Ed Co , J
or
u
n
U.S. Treasury yield curve.

17 ce
i
or
at
n)
eF
4t
l
d
u
e
io
l
I
(
E
J
r
n
ra
t
.
,
di
h
i
r
)
y
Co , Ju e F
po
t
l
)
4
v
d
u
h,
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
75
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
76 Time-Varying Rates of Return and the Yield Curve
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
5.1 Working with Time-Varying Rates of Return
y
d
o
n
01
ul
W
v
4
rp
h,
te
o
7.
a
I
In the real world, rates of return usually differ depending on when the payments are made. For
(
,J
th
2
Interest rates can differ
n
an vo
a
t
o
01
example, the interest rate next year could be higher or lower than it is this year. Moreover, it
.
di
n
based on the length of the
i
lch ion or
C
y
F
is often the case that long-term bonds offer different interest rates than short-term bonds. You
o
.I
commitment.
na
ul
W
E
y2
Iv
must be able to work in such an environment, so let me give you the tools.
p
,
e
o
h
a
ce
J
r
at
th
i
y2
in
o
c
F
t
o
.
),
i
Compounding Different Rates of Return
l
Iv
r
C
d
e
20
F
e
o
l
W
n
E
t
Ju
p
,
n
Fortunately, when working with time-varying interest rates, all the tools you have learned in
a
7.
ch
a
A compounding example with
or
th
ly
Fi
2
or
previous chapters remain applicable (as promised). In particular, compounding still works exactly
in
o
c
t
time-dependent rates of
01

di
l
n
the same way. For example, what is the two-year holding rate of return if the rate of return is 20%
Iv
Ju
,C
ly
4
return.
F
o
na
in the first year and 30% in the second year? (The latter is sometimes called the reinvestment
W
(
oW E
y2
Ju
p
io
rate.) You can determine the two-year holding rate of return from the two 1-year rates of return
,
ce
or
)
h
i
2
r
o
F
using the same compounding formula as before:
t
n
tio po
ul
4t
di
n
Iv
po ly
lch io
e
0
e
o
a
e(
E
2
(1 + r0,2 ) = (1 + 20%) · (1 + 30%) = (1 + 56%)

or
at
Ju
p
in
o
7 .
r
h
ly
i
or
C
(1 + r0,1 ) · (1 + r1,2 ) = (1 + r0,2 )
c
c
t
o
n
01
i
l
n
v
Ju
r
4
d
e
rp
,
Subtract 1, and the answer is a total two-year holding rate of return of 56%. If you prefer it
I
na
W
e(
2
o
a
shorter,
,
i
r
)
ly
i
Ed
iti
r
an vo
c
c
F
o
n
po
l
an
r0,2 = 1.20 · 1.30 – 1 = 1.56 – 1 = 56%
(4
d
lch io
e
20
te
n
h
W
h
or
t
n
elc
(4 elc itio
The calculation is not conceptually more difficult, but the notation is. You have to subscript
a
.
t
di
ce
(4
ly
r
C
not just the interest rates that begin now, but also the interest rates that begin in the future.
c
n
7. e (4 elc itio po
W
l
u
Therefore, most of the examples in this chapter must use two subscripts: one for the time when
d
o
e
0
,
W
(
elc iti
the money is deposited, and one for the time when the money is returned. Thus, r1,2 describes
2
t
n
vo
h
),
ce
an interest rate from time 1 to time 2. Aside from this extra notation, the compounding formula
7
4
i
Ed
r
e
01
is still the very same multiplicative “one-plus formula” for each interest rate (subtracting 1 at the
tio po
l
Fi 7. I
(
W
4
,
end).
e
W
vo th
(
2
r
nc
You can also compound to determine holding rates of return in the future. For example, if
a
o
7.
ce
The general formula for
)
(4
i
r
the 1-year rate of return is 30% from year 1 to year 2, 40% from year 2 to year 3, and 50% from
,C
compounding over many

a
01
o
t
W
year 3 to year 4, then what is your holding rate of return for investing beginning next year for
u
n
periods.
p
e
7.
three years? It is
h
t
c
.
t
1
e
7
4
or
0
Given: r1,2 = 30% r2,3 = 40% r3,4 = 50%
c
F
on
01
I
(
u
y2
p
h,
(1 + r1,4 ) = (1 + 30%) · (1 + 40%) · (1 + 50%) = (1 + 173%)

.
01 nce
J
h
i
7
elc iti
r
F
elc
7.
t
1
,
l
(1 + r1,2 ) · (1 + r2,3 ) · (1 + r3,4 ) = (1 + r1,4 )

)
v
Ju
F
y 2 na
1
I
W
t
Subtracting 1, you see that the three-year holding rate of return for an investment that takes
lch io
0
ra
7.
h
h
i
or
money next year (not today!) and returns money in 4 years (appropriately called r1,4 ) is 173%.
F
t
an vo
o
di
l
Let’s be clear about the timing. For example, say it was midnight of December 31, 2016, right
4
or , Ju e Fi
u
p
,C
e
now. This would be time 0. Time 1 would be midnight December 31, 2017, and this is when
I
(
o W Ed Co , J
or
u
n
you would invest your $1. Three years later, on midnight December 31, 2020 (time 4), you
17 ce
i
or
at
n)
eF
would receive your original dollar plus an additional $1.73, for a total return of $2.73. Interest
4t
l
d
u
rates that begin right now—where the first subscript would be 0—are usually called spot rates.
e
io
l
I
(
E
J
r
Interest rates that begin in the future are usually called forward rates.
n
ra
t
.
,
di
h
i
r
)
y
Co , Ju e F
po
t
l
)
4
v
d
u
h,
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
5.1. Working with Time-Varying Rates of Return 77
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
o
n
01
ul
W
E
Q 5.1. If the first-year interest rate is 2% and the second year interest is 3%, what is the two-year
v
4
rp
h,
te
o
7.
a
I
(
,J
total interest rate?
th
2
n
an vo
a
t
o
01
.
di
n
Q 5.2. Although a two-year project had returned 22% in its first year, overall it lost half of its
i
lch ion or
C
y
F
o
.I
na
ul
W
value. What was the project’s rate of return after the first year?
E
y2
Iv
p
,
e
o
h
a
ce
J
Q 5.3. From the closing of December 31, 2009 to December 31, 2015, Vanguard’s S&P 500 fund
at
th
i
y2
in
o
c
F
t
o
.
(which received and paid dividends on the underlying constituent stocks to its fund investors,
),
i
l
Iv
r
C
d
e
20
F
e
but charged administration fees) returned the following annual rates of return:
o
l
W
n
E
t
Ju
p
,
n
a
7.
ch
a
or
th
ly
Fi
2
or
2010 2011 2012 2013 2014 2015
in
o
c
t
01

di
l
n
Iv
Ju
15.0% 2.1% 16.0% 32.3% 13.7% 1.3%
,C
ly
4
e
F
o
na
W
(
oW E
y2
Ju
p
io
,
What was the rate of return over the first 3 years, and what was it over the second 3 years? What
ce
or
)
h
i
2
r
o
F
t
was the rate of return over the whole 6 years? Was the realized rate of return time-varying?
n
tio po
ul
4t
di
n
Iv
po ly
lch io
e
0
e
o
Q 5.4. A project lost one-third of its value the first year, then gained fifty percent of its value,
a
e(
E
2
or
at
Ju
p
in
o
then lost two-thirds of its value, and finally doubled in value. What was the average rate of
7 .
r
h
ly
i
or
C
return? What was the investment’s overall four-year rate of return? If one is positive, is the
c
c
t
o
n
01
i
l
n
v
Ju
other, too?
r
4
d
e
rp
,
I
na
W
e(
2
o
a

,
i
r
)
ly
i
Ed
iti
r
an vo
c
c
F
o
n
po
Annualized Rates of Return
l
an
(4
d
lch io
e
20
te
n
h
W
h
or
Time-varying rates of return create a new complication that is best explained by an analogy. Is a
t
n
elc
(4 elc itio
a
.
t
Per-unit standard measures

di
ce
car that travels 163,680 yards in 93 minutes fast or slow? It is not easy to say, because you are
(4
ly
r
C
are statistics that are
c
n
7. e (4 elc itio po
used to thinking in “miles per sixty minutes,” not in “yards per ninety-three minutes.” It makes
W
l
conceptual aids.
u
d
o
e
0
,
sense to translate speeds into miles per hour for the purpose of comparing them. You can even
J
W
(
elc iti
2
t
n
vo
do this for sprinters, who run for only 10 seconds. Speeds are just a standard measure of the
a
h
),
ce
7
4
i
Ed
rate of accumulation of distance per unit of time.

e
01
tio po
l
Fi 7. I
(
W
The same issue applies to rates of return: A rate of return of 58.6% over 8.32 years is not as
4
,
e
A per-unit standard for

h
W
vo th
(
easy to compare to other rates of return as a rate of return per year. Therefore, most rates of
2
r
nc
rates of return: annualized

a
o
7.
ce
return are quoted as annualized rates. The average annualized rate of return is just a convenient
)
rates.
(4
i
r
,C
F
a
01
unit of measurement for the rate at which money accumulates—a “sort-of-average” measure of
o
t
W
u
n
performance. Of course, when you compute such an annualized rate of return, you do not mean
e
7.
h
h
t
c
that the investment earned the same annualized rate of return of, say, 5.7% each year—just as
a
.
t
1
e
7
4
the car need not have traveled at 60 mph (163,680 yards in 93 minutes) each instant.
d
or
0
c
F
on
01
I
(
If you were earning a total three-year holding rate of return of 173% over the three-year
u
y2
p
h,
Return to our example: You

.
01 nce
period, what would your annualized rate of return be? The answer is not the average rate of
h
i
7
elc iti
want to annualize our

F
elc
7.
t
return of 173%/3 ≈ 57.7%, because if you earned 57.7% per year, you would have ended up
1
,
l
three-year total holding

)
v
Ju
with 1.5773 – 1 ≈ 292%, not 173%. This incorrect answer of 57.7% ignores the compounded
F
y 2 na
rate of return.
n
1
I
W
t
interest on the interest that you would earn after the first and second years. Instead, to compute
lch io
0
ra
7.
h
h
i
or
the annualized rate of return, you need to find a single hypothetical annual rate of return that, if
F
t
an vo
o
di
l
you received it each and every year, would give you a three-year holding rate of return of 173%.
4
or , Ju e Fi
u
p
,C
e
I
(
How can you compute this? Call this hypothetical annual rate that you would have to earn
o W Ed Co , J
or
u
n
To nd the t-year annualized

17 ce
each year for three years r3 (note the bar above the 3 to denote annualized) in order to end up
i
interest rate, take the t-th

or
at
n)
eF
with a holding rate of return of 173%. To find r3 , solve the equation

4t
root of the total return (t is

l
d
u
number of years).
io
l
I
(
E
J
r
(1 + r3 ) · (1 + r3 ) · (1 + r3 ) = (1 + 173%)
n
ra
t
.
,
di
h
i
r
)
(1 + r3 ) · (1 + r3 ) · (1 + r3 ) = (1 + r0,3 )
Co , Ju e F
po
t
l
)
4
v
d
u
h,
or, for short,

a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
o
n
01
(1 + r3 )3 = (1 + 173%)
ul
W
v
4
rp
h,
te
o
7.
a
I
(
,J
(1 + rt )t =
th
i
(1 + r0,t ) (5.1)
2
n
an vo
a
t
o
01
.
di
n
i
lch ion or
C
y
In our example, the holding rate of return r0,3 is known (173%) and the annualized rate of
o
.I
na
ul
W
E
y2
Iv
return r3 is unknown. Earning the same rate (r3 ) three years in a row should result in a holding
p
,
e
o
h
a
ce
J
r
at
th
rate of return of 173%. It is a “smoothed-out” rate of return of the three years’ rates of return.
i
y2
in
o
c
F
t
o
.
),
Think of it as a hypothetical, single, constant-speed rate at which your money would have ended
i
l
Iv
r
C
d
e
20
F
e
o
up as quickly at 173% as it did with the 30%, 40%, and 50% individual annual rates of return.
l
W
n
E
t
Ju
p
,
n
The correct solution for r3 is obtained by computing the third root of 1 plus the total holding
a
7.
ch
a
or
th
ly
Fi
2
or
in
rate of return:
o
c
t
01

di
l
n
Iv
Ju
,C
ly
p
4
e
F
o
(1 + r3 ) = (1 + 173%)( /3) =
3
na
ä Math Background,
1 + 173% ≈ 1 + 39.76%
W
(
oW E
y2
Ju
Chapter A, Pg.621.
p
io
Æ
,
ce
(1/t)
or
)
(1 + r0,t ) t
1 + r0,t = (1 + rt )
h
i
2
r
=
o
F
t
n
tio po
ul
4t
di
n
Iv
po ly
Confirm with your calculator that r3 ≈ 39.76%,
lch io
e
0
e
o
a
e(
E
2
or
at
Ju
p
in
o
.
1.3976 · 1.3976 · 1.3976 ≈ (1 + 173%)

7
r
h
ly
i
or
C
c
c
t
o
n
01
i
l
(1 + r3 ) · (1 + r3 ) · (1 + r3 ) = (1 + r0,3 )
n
v
Ju
r
4
d
e
rp
,
I
na
W
e(
2
In sum, if you invested money at a rate of 39.76% per annum for 3 years, you would end up
o
a

,
i
r
)
ly
with a total three-year holding rate of return of 173%. As is the case here, for very long periods,
Ed
iti
r
an vo
c
c
F
o
n
po
the order of magnitude of the annualized rate will often be so different from the holding rate
l
an
(4
d
lch io
e
20
te
n
h
that you will intuitively immediately register whether the quantity r0,3 or r3 is meant. In the
W
h
or
t
n
elc
(4 elc itio
a
real world, very few rates of return, especially over long horizons, are quoted as holding rates of
.
t
di
ce
(4
ly
r
C
return. Almost all rates are quoted in annualized terms instead.
c
n
7. e (4 elc itio po
W
l
u
d
o
e
0
,
IMPORTANT
J
W
(
elc iti
2
t
n
vo
The total holding rate of return over t years, called r0,t , is translated into an annualized rate of
a
h
),
ce
7
4
i
Ed
return, called rt , by taking the t-th root:

e
01
tio po
l
Fi 7. I
(
W
4
Æ
,
e
1 + r0,t = (1 + r0,t )1/t

h
(1 + rt ) =
W
t
vo th
(
2
r
nc
a
o
7.
ce
)
(4
i
r
,C
Compounding the annualized rate of return over t years yields the total holding rate of return.
F
a
01
o
t
W
u
n
e
7.
h
h
t
c
.
You also will often need to compute annualized rates of return from payoffs yourself. For
t
1
e
7
4
Translating long-term dollar

d
or
0
example, what annualized rate of return would you expect from a $100 investment today that
c
F
on
01
I
returns into annualized

(
u
y2
promises a return of $240 in 30 years? The first step is computing the total holding rate of return.
p
h,
rates of return.
.
01 nce
J
h
i
7
elc iti
Take the ending value ($240) minus your beginning value ($100), and divide by the beginning
r
F
elc
7.
t
1
,
l
value. Thus, the total 30-year holding rate of return is

)
v
Ju
F
y 2 na
1
I
$240 – $100
t
u
lch io
0
ra
r0,30 = = 140%
7.
$100
h
i
or
F
t
an vo
o
di
l
C30 – C0
4
or , Ju e Fi
u
p
,C
r0,30 =
e
C0
(
o W Ed Co , J
or
u
n
17 ce
The annualized rate of return is the rate r30 , which, if compounded for 30 years, offers a 140%
i
or
at
n)
eF
4t
rate of return,
l
d
u
e
io
l
I
(
E
J
r
(1 + r30 )30 = (1 + 140%)

t
n
ra
t
.
,
di
h
i
r
)
(1 + rt )t
y
Co , Ju e F
= (1 + r0,t )
po
t
l
)
4
v
d
u
h,
Solve this by taking the 30th root,

a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
p
o
n
01
(1 + r30 ) = (1 + 140%)1/30 =
30
ul
1 + 140% ≈ 1 + 2.96% W
v
4
rp
h,
te
o
7.
a
I
(
,J
1/t
th
i
(1 + rt ) (1 + r0,t ) 1 + r0,t
2
t
= =
n
an vo
a
t
o
01
.
di
n
i
lch ion or
C
y
Subtracting 1, you see that a return of $240 in 30 years for an initial $100 investment is equivalent
o
.I
na
ul
W
E
y2
Iv
to a 2.96% annualized rate of return.
p
,
e
o
h
a
ce
J
r
at
th
i
y2
In the context of rates of return, compounding is similar to adding, while annualizing is
in
o
c
F
t
o
Compounding ≈ adding.
.
),
i
l
similar to averaging. If you earn 1% twice, your compounded rate is 2.01%, similar to the rates
Iv
r
C
d
e
20
Annualizing ≈ averaging.
F
e
o
themselves added (2%). Your annualized rate of return is 1%, similar to the average rate of
l
W
n
E
t
Ju
p
,
n
a
7.
return of 2.01%/2 = 1.005%. The difference is the interest on the interest.
ch
a
or
th
ly
Fi
2
or
in
o
c
t
Now assume that you have an investment that doubles in value in the first year and then falls
01

di
l
n
Iv
Averaging can lead to
Ju
,C
ly
4
e
back to its original value. What would its average rate of return be? Doubling from, say, $100 to
F
o
na
surprising results—returns
W
(
oW E
y2
Ju
$200 is a rate of return of +100%. Falling back to $100 is a rate of return of ($100–$200)/$200 =
p
that are much higher than
io
,
ce
or
–50%. Therefore, the average rate of return would be [+100% + (–50%)]/2 = +25%.
)
h
i
what you earned per year.
2
r
o
F
t
n
tio po
ul
4t
di
But you have not made any money! You started with $100 and ended up with $100. If you
n
Iv
po ly
lch io
e
0
e
o
a
compound the returns, you get the answer of 0% that you were intuitively expecting: e(
E
2
or
at
Ju
p
in
o
7 .
r
Look how deceptive!
h
ly
i
or
(1 + 100%) · (1 – 50%) = 1 + 0% ⇒ r0,2 = 0%

C
c
c
t
o
n
01
i
l
n
v
Ju
r
4
d
e
rp
,
(1 + r0,1 ) · (1 + r1,2 ) = (1 + r0,2 )

I
na
W
e(
2
o
a

It follows that the annualized rate of return r2 is also 0%. Conversely, an investment that produces
,
i
r
)
ly
i
Ed
iti
r
an vo
c
c
F
o
+20% followed by –20% has an average rate of return of 0% but leaves you with a loss:
n
po
l
an
(4
d
lch io
e
20
te
n
h
W
h
(1 + 20%) · (1 – 20%) = (1 – 4%) ⇒ r0,2 = –4%

or
t
n
elc
(4 elc itio
a
.
t
di
ce
(4
ly
r
(1 + r0,1 ) · (1 + r1,2 ) = (1 + r0,2 )

C
c
n
7. e (4 elc itio po
W
l
u
d
o
e
0
,
For every $100 of your original investment, you now have only $96. The average rate of return
J
W
(
elc iti
2
t
of 0% does not reflect this loss. Both the compounded and therefore the annualized rates of
n
vo
h
),
ce
7
4
return do tell you that you had a loss:

Ed
r
e
01
tio po
l
Fi 7. I
(
W
Æ p
4
,
1 + r2 = (1 + r0,2 ) = 1 – 4% = 1 – 2.02% ⇒ r2 ≈ –2.02%

e
ä More about how arithmetic returns

h
W
vo th
(
2
r
are “too high” in normally-distributed

nc
a
o
7.
If you were an investment advisor and quoting your historical performance, would you rather stock returns,
ce
)
(4
i
r
,C
Pg.141.
F
a
01
quote your average historical rate of return or your annualized rate of return? (Hint: The
o
t
W
u
n
industry standard is to quote the average rate of return, not the annualized rate of return!)
e
7.
h
h
t
c
Make sure to solve the following questions to gain more experience with compounding and
a
.
t
1
e
7 Do it!
4
or
annualizing over different time horizons.

0
c
F
on
01
I
(
u
y2
p
h,

.
01 nce
J
h
i
7
elc iti
Q 5.5. If you earn a rate of return of 5% over 4 months, what is the annualized rate of return?
F
elc
7.
t
1
,
l
)
v
Ju
Q 5.6. Assume that the two-year holding rate of return is 40%. The average (arithmetic) rate
F
y 2 na
1
I
W
t
of return is therefore 20% per year. What is the annualized (geometric) rate of return? Is the
lch io
0
ra
7.
h
h
i
annualized rate the same as the average rate?

or
F
t
an vo
o
di
l
Q 5.7. Is the compounded rate of return higher or lower than the sum of the individual rates of
or , Ju e Fi
u
p
,C
e
I
(
o W Ed Co , J
or
return? Is the annualized rate of return higher or lower than the average of the individual rates
t
u
n
17 ce
of return? Why?
i
or
at
n)
eF
4t
l
Q 5.8. Return to Question 5.3. What was the annualized rate of return on the S&P 500 over the
v
d
u
e
io
six years in the table?

l
I
(
E
J
r
n
ra
t
Q 5.9. If the total holding interest rate is 50% for a five-year investment, what is the annualized
,
di
h
i
r
)
y
Co , Ju e F
po
rate of return?
t
l
)
4
v
d
u
h,
0
a
Q 5.10. If the per-year interest rate is 10% for each of the next 5 years, what is the annualized
I
e(
J
or
W
y2
n
c
five-year rate of return?

7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
Duration and Maturity
y
d
o
n
01
ul
W
v
4
rp
h,
te
o
7.
We sometimes need summary statistics of how long a bond or a project will last. Maturity is
a
I
(
,J
th
2
n
an vo
the date of the very last payment of a bond. However, you would probably not consider a bond
a
t
o
01
.
di
n
i
lch ion or
C
y
that pays $1 in one year and $1 in 30 years a 30-year bond—it’s more like a 15-year bond. The
o
.I
na
ul
W
E
y2
duration can be calculated as
Iv
p
,
e
o
h
a
ce
J
r
at
th
i
y2
1 × $1 + 2 × $0 + . . . + 29 × $0 + 30 × $1
in
o
c
F
t
o
.
Duration( C1 = $1, C30 = $1 ) = = 15.5 years
),
i
l
Iv
r
C
$1X
+ $0 + . . . + $1
d
e
20
F
e
o
l
W
n
E
t
Duration( {Ct } ) t/W × Ct /
Ju
=
p
,
n
a
7.
ch
a
t
or
th
ly
Fi
2
or
in
o
c
t
01

di
l
n
Iv
Ju
where W is the sum of all payments, W = Ct . Intuitively, this bond has about a 15-year
,C
ly
4
e
F
o
na
W
(
oW E
y2
Ju
p
io
duration. Also intuitively, a bond that pays $100 in one year and $1 in 30 years has a shorter
,
ce
or
)
h
i
2
r
duration,
o
F
t
n
tio po
ul
4t
di
n
Iv
po ly
lch io
e
0
e
1 × $100 + 30 × $1
o
a
e(
E
2
or
Duration( C1 = $100, C30 = $1 ) = ≈ 1.287

at
Ju
p
in
o
$100 + $1
7 .
r
h
ly
i
or
C
c
c
t
o
n
01
A zero bond has a duration equal to its maturity. One important variation is the Macauley
i
l
n
v
Ju
r
4
d
e
rp
,
Duration, which essentially replaces the raw cash flow with its present value. The intent is to
I
na
W
e(
2
discount far-away payouts more, which thus further shortens duration. For example, using our
o
a

,
i
r
)
ly
i
Ed
iti
December 2015 yield curve,
r
an vo
c
c
F
o
n
po
l
an
(4
d
lch io
e
20
1 × $100/1.0065 + 99 × $1/1.030130
te
n
h
W
h
Duration( C1 = $1, C100 = $1 ) = ≈ 1.004

or
t
$100/1.0065 + $1/1.030130
n
elc
(4 elc itio
a
.
t
di
ce
(4
ly
r
C
c
There are also other measures of duration, but they are all basically summary statistics of when
n
7. e (4 elc itio po
W
l
u
d
o
e
0
,
your “average” cash flow is going to occur.

J
W
(
elc iti
2
t
n
vo
h
),
ce
7
4
Present Values with Time-Varying Interest Rates

Ed
r
e
01
tio po
l
Fi 7. I
(
W
4
,
Let’s proceed now to net present value with time-varying interest rates. What do you need
e
W
vo th
(
The PV formula still looks
2
r
nc
to learn about the role of time-varying interest rates when computing NPV? The answer is
a
o
very similar.
7.
ce
)
(4
i
r
essentially nothing new. You already know everything you need to know here. The net present
,C
F
a
01
o
t
W
value formula is still

u
n
e
7.
h
h
t
c
.
t
1

i
e
7
4
NPV = PV C0 + PV C1 PV C2 PV C3
d
or
+ + + ···
0
c
F
on
01
I
(
u
y2
p
h,
C1 C2 C3
.
01 nce
J
h
i
C0 + ···
7
elc iti
= + + +
F
1 + r0,1 1 + r0,2 1 + r0,3

elc
7.
t
1
,
l
)
v
Ju
C1 C2 C3
e
F
y 2 na
1 C0 + ···
I
= + + +
W
t
(1 + r2 )2 (1 + r3 )3
u
1 + r1
lch io
0
ra
7.
h
h
i
C1 C2 C3
or
F
C0 + ···
an vo
o
= + + +
t
di
l
1 + r0,1 (1 + r0,1 ) · (1 + r1,2 ) (1 + r0,1 ) · (1 + r1,2 ) · (1 + r2,3 )

4
or , Ju e Fi
u
p
,C
e
I
(
o W Ed Co , J
or
The only novelty is that you need to be more careful with your subscripts. You cannot simply
n
17 ce
i
assume that the multiyear holding returns (e.g., 1 + r0,2 ) are the squared 1-year rates of return
or
at
n)
eF
4t
((1 + r0,1 )2 ). Instead, you must work with time-dependent costs of capital (interest rates).
l
d
u
e
io
That’s it.
I
(
E
J
r
n
ra
t
For example, say you have a project with an initial investment of $12 that pays $10 in one
,
di
h
i
r
)
Present values are still alike

y
Co , Ju e F
po
year and $8 in five years. Assume that the 1-year interest rate is 5% and the five-year annualized
t
l
and thus can be added,

)
4
v
d
u
h,
interest rate is 6% per annum. In this case,

0
subtracted, compared, and

a
n
I
e(
J
or
W
y2
so on.
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d

o
n
$10
01
ul
W
E
PV $10 in 1 year
v
≈ $9.52
4
rp
=
h,
te
o
7.
1.05
a
I
(
,J
th
2
n
an vo
a
t
o

01
$8
.
di
n
PV $8 in 5 years
i
≈ $5.98
lch ion or
C
y
=
o
1.065
.I
na
ul
W
E
y2
Iv
p
,
e
o
It follows that the project’s total value today (time 0) is $15.50. If the project costs $12, its net
a
ce
J
r
at
th
i
y2
in
o
c
F
t
o
present value is
.
),
i
l
Iv
r
C
d
e
20
F
e
o
$10 $8
l
W
n
E
t
NPV = – $12 + ≈ $3.50
Ju
p
+
,
n
a
7.
1.05 1.065
ch
a
or
th
ly
Fi
2
or
in
o
c
t
C5
01
C1

di
l
n
Iv
Ju
NPV = C0 = NPV
,C
ly
4
+ +
F
1 + r0,1 1 + r0,5
o
na
W
(
oW E
y2
Ju
p
io
,
ce
or
)
h
i
2
r
You can also rework a more involved project, similar to that in Exhibit 2.3,. But to make
o
F
t
n
tio po
ul
4t
Here is a typical NPV
di
n
Iv
it more interesting, let’s now use a hypothetical current term structure of interest rates that is
po ly
lch io
e
0
e
example.
o
a
e(
E
upward-sloping. Assume this project requires an appropriate annualized discount rate of 5%
2
or
at
Ju
p
in
o
.
over one year, and 0.5% more for every subsequent year, so that the cost of capital reaches 7%
7
r
h
ly
i
or
C
ä Hypothetical Project Cash Flows,
c
c
t
o
n
01
annualized in the fifth year. The valuation method works the same way as it did in Exhibit 2.3—
t
Exhibit 2.3, Pg.28.
i
l
n
v
Ju
r
4
d
e
rp
,
you only have to be a little more careful with the interest rate subscripts. The project’s value is
I
na
W
e(
2
thus
o
a

,
i
r
)
ly
i
Ed
iti
r
an vo
c
c
F
o
n
po
Project Annualized Com- Discount
l
an
(4
d
lch io
e
20
te
Time Cash Flow Rate pounded Factor I Value
n
h
W
h
or
t

elc
(4 elc itio
a
1
.
t
di
ce PV Ct
t Ct rt r0,t
(4
ly
r
C
c
1 + r0,t
n
7. e (4 elc itio po
W
l
u
d
o
e
0
,
Today –$900 any 1.0000 –$900.00

J
W
(
elc iti
2
t
n
vo
h
),
Y1 5.0% 5.0% 0.9524 $190.48
ce
+$200
7
4
i
Ed
r
e
Y2 5.5% 11.3% 0.8985 $179.69

n
01
tio po
+$200
l
Fi 7. I
(
W
4
Y3 6.0% 19.1% 0.8396 $335.85
,
+$400
e
W
vo th
(
2
r
Y4 6.5% 28.6% 0.7773 $310.93

nc
+$400
a
o
7.
ce
Y5 –$100 7.0% 40.3% 0.7130 –$71.30
)
(4
i
r
,C
F
a
01
o
t
W
Net Present Value (Sum): $45.65

n
e
7.
h
h
t
c
.
t
1
e
7
4
or
Q 5.11. A project costs $200 and will provide cash flows of +$100, +$300, and +$500 in
0
c
F
on
01
I
(
u
y2
consecutive years. The annualized interest rate is 3% per annum over one year, 4% per annum
p
h,

.
01 nce
J
h
i
7
over two years, and 4.5% per annum over three years. What is this project’s NPV?
elc iti
r
F
elc
7.
t
1
,
l
)
v
Ju
F
y 2 na
1
I
W
t
u
lch io
0
ra
7.
h
h
i
or
F
t
an vo
o
di
l
or , Ju e Fi
u
p
,C
e
I
(
o W Ed Co , J
or
u
n
17 ce
i
or
at
n)
eF
4t
l
d
u
e
io
l
I
(
E
J
r
n
ra
t
.
,
di
h
i
r
)
y
Co , Ju e F
po
t
l
)
4
v
d
u
h,
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
5.2 Inflation
y
d
o
n
01
ul
W
v
4
rp
h,
te
o
7.
a
I
Let’s make our world even more realistic—and complex—by working out the effects of inflation.
(
,J
th
2
In ation is the increase in
n
an vo
a
t
o
01
Inflation is the process by which the same good costs more in the future than it does today. With
.
di
n
the price of the same good.
i
lch ion or
C
y
F
inflation, the price level is rising and thus money is losing its value. For example, if inflation is
o
.I
na
ul
W
E
y2
Iv
100%, an apple that costs $0.50 today will cost $1 next year, a banana that costs $2 today will
p
,
e
o
h
a
ce
J
r
at
th
i
cost $4, and corporate finance textbooks that cost $200 today will cost $400.
y2
in
o
c
F
t
o
.
),
i
l
Iv
Inflation may or may not matter in a corporate context, depending on how contracts are
r
C
d
e
20
F
e
In ation matters when
o
l
W
written. If you ignore inflation and write a contract that promises to deliver bread for the price
n
E
t
contracts are not written to
Ju
p
,
n
a
7.
of $1 next year, it is said to be in nominal terms—and you may have made a big mistake. The
ch
a
adjust for it.
or
th
ly
Fi
2
or
in
o
c
money you will be paid will be worth only half as much. You will only be able to buy one apple
t
01

di
l
n
Iv
Ju
,C
ly
4
for each loaf of bread that you had agreed to sell for $1, not the two apples that anyone else will
F
o
na
W
(
oW E
y2
enjoy. On the other hand, you could write your contract in real terms (or inflation-indexed
Ju
p
io
,
ce
terms) today, in which case the inflationary price change would not matter. That is, you could
or
)
h
i
2
r
o
F
t
n
tio po
ul
build into your promised banana delivery price the inflation rate from today to next year. An
4t
di
n
Iv
po ly
lch io
e
0
e
example would be a contract that promises to deliver bananas at the rate of four apples per
o
a
e(
E
2
or
at
Ju
p
banana. If a contract is indexed to inflation, then inflation does not matter. However, in the
in
o
7 .
r
h
ly
i
or
United States inflation often does matter, because most contracts are in nominal terms and not
C
c
c
t
o
n
01
i
l
inflation-indexed. Therefore, you have to learn how to work with inflation. What effect, then,
n
v
Ju
r
4
d
e
rp
,
I
na
does inflation have on returns? On (net) present values? This is our next subject.
W
e(
2
o
a

,
i
r
)
ly
i
Ed
iti
r
an vo
c
c
F
Measuring the Inflation Rate
o
n
po
l
an
(4
d
lch io
e
20
te
n
h
W
The first important question is how you should define the inflation rate. Is the rate of change
h
or
t
The CPI is the most common

elc
(4 elc itio
a
.
t
of the price of apples the best measure of inflation? What if apples (the fruit) become more
di
ce
in ation measure.
(4
ly
r
C
c
expensive, but Apples (the computers) become less expensive? Defining inflation is actually
n
7. e (4 elc itio po
W
l
u
d
o
e
0
rather tricky. To solve this problem, economists have invented baskets or bundles of goods
,
W
(
elc iti
2
t
and services that are deemed to be representative. Economists then measure an average price
n
vo
h
),
ce
7
change for these items. The official source of most inflation measures is the Bureau of Labor
4
i
Ed
r
e
01
tio po
l
Fi 7. I
Statistics (BLS), which determines the compositions of a number of common bundles (indexes)
W
4
,
and publishes the average total price of these bundles on a monthly basis. The most prominent
e
W
vo th
(
2
r
nc
such inflation measure is a hypothetical bundle of average household consumption, called the
a
o
7.
ce
)
(4
i
r
,C
Consumer Price Index (CPI). (The CPI components are roughly: housing 40%, food 20%,
F
a
01
o
t
W
transportation 15%, medical care 10%, clothing 5%, entertainment 5%, others 5%.) The BLS
u
n
e
7.
offers inflation data at http://www.bls.gov/cpi/, and the Wall Street Journal prints the percent
h
t
c
.
t
change in the CPI at the end of its regular column “Money Rates.”
1
e
7
4
or
0
c
F
on
From Dec 2014 to Dec 2015, the inflation rate was a remarkably low 0.7% per annum. (And
01
I
(
u
y2
p
h,
there were some months with negative inflation rates, too—called deflation!)
.
01 nce
J
h
i
7
elc iti
r
F
elc
7.
t
1
,
l
Year 2010 2011 2012 2013 2014 2015

)
v
Ju
F
y 2 na
1
I
W
t
CPI 1.5 3.0 1.7 1.5 0.8 0.7

lch io
0
ra
7.
h
h
i
or
F
t
an vo
o
A number of other indexes are also commonly used as inflation measures, such as the
di
l
or , Ju e Fi
u
p
,C
e
Producer Price Index (PPI) or the broader GDP Deflator. They typically move very similarly
I
(
o W Ed Co , J
or
u
n
to the CPI. Over short periods, one expects these rates to move fairly close together (on average,
17 ce
i
or
not every month); but over longer periods, they can diverge. There are also more specialized
at
n)
eF
4t
l
bundles, such as computer or flash memory chip inflation indexes (their prices usually decline),
v
d
u
e
io
or price indexes for particular regions.

I
(
E
J
r
n
ra
t
The official inflation rate is not just a number that mirrors reality—it is important in itself,
,
di
h
i
The CPI matters—even if it

)
y
Co , Ju e F
because many contracts are specifically indexed to a particular inflation definition. For example,
po
t
l
is calculated incorrectly.
)
4
v
d
u
h,
even if actual true inflation is zero, if the officially reported CPI rate is positive, the government
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
5.2. Inflation 83
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
o
The German Hyperin ation of 1922
01
ul
W
v
4
rp
h,
te
o
7.
a
I
(
,J
th
Many economists now believe that a modest inflation rate between 1% and 3% per year is a healthy number. It’s not so
2
n
an vo
a
t
o
01
.
easy to maintain.
di
n
i
lch ion or
C
y
F
o
.I
na
ul
W
The Roman Emperor Gallienus is still infamous two-thousand years later for having changed the value of the Roman
E
y2
Iv
p
,
e
o
h
a
Denarius by reducing its silver content by a factor of 100.
ce
J
r
at
th
i
y2
in
o
c
F
t
o
.
The most famous episode of extreme inflation (hyperinflation) occurred in Germany from August 1922 to November
),
i
l
Iv
r
C
d
e
20
F
e
1923. Prices more than quadrupled every month. The price for goods was higher in the evening than in the morning!
o
l
W
n
E
t
Ju
p
,
n
Stamps had to be overprinted by the day, and shoppers went out in the morning with bags of money that were worthless
a
7.
ch
a
or
th
ly
Fi
2
or
by the end of the day. By the time Germany printed 1,000 billion Mark Bank Notes, no one trusted Marks anymore. This
in
o
c
t
01

di
l
n
hyperinflation was stopped only by a drastic currency and financial system reform. But high inflation is not just a historic
Iv
Ju
,C
ly
4
e
F
o
na
artifact. In 2015, Venezuela had an inflation rate of 180%. Don’t trust Venezuelan Bolivars!
W
(
oW E
y2
Ju
p
io
,
ce
Yet recent experience has humbled us economists (further) by proving that it can also be difficult to push up inflation. In
or
)
h
i
2
r
o
F
t
n
the Great Recession (the financial crisis of 2008-11), the Fed tried to fuel inflation by pushing money into the hands of
tio po
ul
4t
di
n
Iv
po ly
lch io
e
0
consumers. The idea was to get them to lose just a little trust in the currency and spend it, so as to raise the value of
e
o
a
e(
E
2
or
at
Ju
much underwater real estate. But consumers turned around and deposited the money back into their banks, which in turn
p
in
o
7 .
r
h
redeposited the money with—you guessed it—the Fed.
ly
i
or
C
c
c
t
o
n
01
i
l
n
v
Ju
r
4
d
e
rp
,
I
na
W
e(
2
o
a

,
i
r
)
ly
i
Ed
iti
r
must pay out more to Social Security recipients. The lower the official inflation rate, the less
an vo
c
c
F
o
n
po
l
an
the government has to pay. You would therefore think that the government has the incentive to
(4
d
lch io
e
20
te
n
h
understate inflation. But, strangely, this has not been the case. On the contrary, there are strong
W
h
or
t
n
elc
(4 elc itio
a
.
political interest groups that hinder the BLS from even fixing mistakes that everyone knows
t
di
ce
(4
ly
r
C
c
overstate the CPI—that is, corrections that would result in lower official inflation numbers. In
n
7. e (4 elc itio po
W
l
u
4
1996, the Boskin Commission, consisting of a number of eminent economists, found that the CPI
d
o
e
0
,
W
(
elc iti
overstates inflation by about 74 basis points per annum—a huge difference. The main reasons
t
n
vo
h
),
ce
were and continue to be that the BLS has been tardy in recognizing the growing importance of
7
4
i
Ed
r
e
01
tio po
such factors as effective price declines in the computer and telecommunications industries and
l
Fi 7. I
(
W
4
,
the role of superstores such as Wal-Mart and Target.

e
W
vo th
(
2
r
nc
Before we get moving, a final warning:

a
o
7.
ce
)
(4
i
r
,C
F
a
01
o
t
IMPORTANT
W
u
n
e
7.
The common statement “in today’s dollars” can be ambiguous. Most people mean “inflation-
e
h
h
t
c
adjusted.” Some people mean present values (i.e., “compared to an investment in risk-free
a
.
t
1
e
7
4
or
bonds”). When in doubt, ask!

0
c
F
on
01
I
(
u
y2
p
h,

.
01 nce
J
h
i
7
elc iti
r
F
elc
7.
t
1
,
l
)
v
Ju
F
y 2 na
1
I
Q 5.12. Read the Bureau of Labor Statistics’ website descriptions of the CPI and the PPI. How
W
t
u
lch io
0
ra
does the CPI differ conceptually from the PPI? Are the two official rates different right now?
7.
h
h
i
or
F
t
an vo
o
di
l
or , Ju e Fi
u
p
,C
e
I
(
o W Ed Co , J
or
u
n
Real and Nominal Interest Rates

17 ce
i
or
at
n)
eF
4t
To work with inflation and to learn how you would properly index a contract for inflation, you
l
d
u
Nominal returns are what is

e
io
first need to learn the difference between a nominal return and a real return. The nominal rate
n
l
I
(
E
J
r
normally quoted. Real

t
n
ra
is what is usually quoted—a return that has not been adjusted for inflation. In contrast, the real
t
returns are adjusted for

.
,
di
h
i
r
)
rate of return “somehow takes out” inflation from the nominal rate in order to calculate a return in ation. They are what you
Co , Ju e F
po
t
l
“as if” there had been no price inflation to begin with. The real return reflects the fact that, in want to know if you want to
)
4
v
d
u
h,
0
a
consume.
I
e(
the presence of inflation, dollars in the future will have less purchasing power than dollars today.
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
It measures your trade-off between present and future consumption, taking into account the
o
n
01
ul
W
v
change in prices.
4
rp
h,
te
o
7.
a
I
(
,J
th
Start with a simple exaggerated scenario: Assume that the inflation rate is 100% per year
2
n
an vo
a
t
o
An extreme 100% in ation
01
.
di
and you can buy a bond that promises a nominal interest rate of 700%. What is your real rate
n
i
lch ion or
C
y
rate example: Prices double
o
.I
na
n
of return? To find out, assume that $1 buys one apple today. With an inflation rate of 100%,
ul
W
every year.
E
y2
Iv
p
,
e
o
you need $2 next year to buy the same apple. Your gross return will be $1 · (1 + 700%) = $8
a
ce
J
r
at
th
i
y2
in
o
for today’s $1 of investment. But this $8 will apply to apples costing $2 each. Your $8 will buy
c
F
t
o
.
),
i
l
Iv
r
C
4 apples, not 8 apples. Your real rate of return (1 apple yields 4 apples) is therefore
d
e
20
F
e
o
l
W
n
E
t
Ju
p
,
n
(4 Apples for $8) – (1 Apple for $2)

a
7.
ch
a
rreal = = 300%
or
th
ly
Fi
2
or
in
(1 Apple for $2)
o
c
t
01

di
l
n
Iv
Ju
,C
ly
4
e
F
For each dollar invested today, you will be able to buy only 300% more apples next year (not
o
na
W
(
oW E
y2
Ju
p
700% more) than you could buy today. This is because inflation will reduce the purchasing
io
,
ce
or
)
h
i
2
r
power of your dollar by more than one half.

o
F
t
n
tio po
ul
4t
di
n
Iv
The correct formula to adjust for the inflation rate (π) is again a “one-plus” type formula. In
po ly
lch io
e
0
e
Here is the correct
o
a
our example, it is e(
E
2
or
at
Ju
conversion formula from
p
in
o
7 .
nominal to real rates.
r
h
ly
i
or
C
c
c
t
o
(1 + 700%) = (1 + 300%) · (1 + 100%)
n
01
i
l
n
v
Ju
r
4
d
e
rp
,
I
na
(1 + rnominal ) = (1 + rreal ) · (1 + π)
W
e(
2
o
a

,
i
r
)
ly
i
Ed
iti
Turning this formula around gives you the real rate of return,
r
an vo
c
c
F
o
n
po
l
an
(4
d
lch io
e
20
1 + 700%
te
n
h
(1 + rreal ) = = 1 + 300%
W
h
or
1 + 100%
t
n
elc
(4 elc itio
a
.
t
di
ce
(4
ly
r
C
(1 + rnominal )
c
n
7. e (4 elc itio po
(1 + rreal ) =
W
l
u
4
(1 + π)
d
o
e
0
,
W
(
elc iti
2
t
In plain English, a nominal interest rate of 700% is the same as a real interest rate of 300%,
vo
h
),
ce
7
4
i
Ed
given an inflation rate of 100%.

e
01
tio po
l
Fi 7. I
(
W
4
,
e
IMPORTANT
h
W
vo th
(
2
r
nc
The relation between the nominal rate of return (rnominal ), the real rate of return (rreal ), and the
a
o
7.
ce
)
(4
i
r
,C
inflation rate (π) is

F
a
01
o
t
W
u
n
e
7.
(1 + rnominal ) = (1 + rreal ) · (1 + π) (5.2)

h
t
c
.
t
1
e
7
4
or
0
c
F
on
01
I
(
u
y2
p
h,

.
01 nce
J
h
i
7
elc iti
r
F
As with compounding, if the rates are small, the mistake of just subtracting the inflation rate
elc
7.
t
1
,
l
For small rates, adding/

)
v
Ju
from the nominal interest rate to obtain the real interest rate is not too grave. For example, with
e
F
subtracting is an okay
y 2 na
1
I
W
t
our (30-year) 3% Treasury, if inflation were to remain 0.7%, the correct real interest rate would
u
approximation.
lch io
0
ra
7.
be 2.28% and not 2.30%:

h
i
or
F
t
an vo
o
di
l
or , Ju e Fi
u
ä Bills, notes, and bonds,

p
,C
e
Sect. 5.3, Pg.86. (1 + 3%) ≈ (1 + 2.28...%) · (1 + 0.7%) ≈ 1 + 2.28...% + 0.7% + 0.0002... . . .

I
(
o W Ed Co , J
or
u
n
(1 + rnominal ) = (1 + rreal ) (1 + π) = 1 + rreal rreal · π

17 ce
· + π +
i
or
| {z }
at
n)
eF
4t
l
cross-term
v
d
u
e
io
l
I
The difference between the correct and the approximation, i.e., the cross-term, is trivial, and
(
E
J
r
n
ra
t
easily swamped by measurement noise in the current inflation rate and uncertainty about future
,
di
h
i
r
)
ä Adding or compounding interest

y
Co , Ju e F
inflation rates. However, when inflation and interest rates are high—as they were, for example,
po
rates, and the cross-term,

l
)
4
v
d
u
h,
Pg.20. in the United States in the late 1970s—then the cross-term can be important.
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
5.2. Inflation 85
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
A positive time value of money—that the same amount of money today is worth more than
o
n
01
Real interest rates can be
ul
W
v
tomorrow—is only true for nominal quantities, not for real quantities. Only nominal interest
4
rp
h,
te
o
negative.
7.
a
I
(
,J
rates are (usually) not negative. In the presence of inflation, real interest rates not only can be
th
2
n
an vo
a
t
o
01
.
negative, but often are negative. For example, in December 2015, the one-month Treasury paid
di
n
i
lch ion or
C
y
F
o
0.14% while the inflation rate was 0.7%, implying a real interest rate of about –0.56%. Every
.I
na
ul
W
E
y2
Iv
p
,
dollar you invested in such U.S. Treasuries would be worth less in real purchasing power one
e
o
h
a
ce
J
r
at
th
i
y2
month later. You would have ended up with more cash—but also with less purchasing power.
in
o
c
F
t
o
.
),
i
l
Iv
r
C
d
e
20
F
e
o
l
W
n
E
t
Ju
Gold
p
,
n
a
7.
ch
a
or
th
ly
Fi
2
or
in
o
c
t
Sometimes, the price of gold has been used as a measure of inflation. Although gold is not a great measure of purchasing
01

di
l
n
Iv
Ju
,C
ly
4
e
power in general, it does make it easy to conduct long-run comparisons. A Roman legionary was paid the equivalent of 2.31
F
o
na
W
(
oW E
y2
Ju
oz of gold a year. A U.S. Army private nowadays is paid the equivalent of 11.01 oz—about five times as much. A Roman
p
io
,
ce
or
centurion was paid 38.58 oz. A U.S. army captain is paid 27.8 oz, about a quarter less. Thus, as an asset class, an investment
)
h
i
2
r
o
F
t
n
tio po
ul
4t
in gold would have earned a rate of return roughly in line with the income growth over two millenia.
di
Erb and Harvey (2013)
n
Iv
po ly
lch io
e
0
e
o
a
e(
E
2
or
at
Ju
p
in
o
7 .
r
h
ly
i
or
C
c
c
t
o
n
01
i
l
n
v
Ju
r
4
d
e
rp
,
Q 5.13. From memory, write down the relationship between nominal rates of return (rnominal ),
I
na
W
e(
2
real rates of return (rreal ), and the inflation rate (π).
o
a

,
i
r
)
ly
i
Ed
iti
r
an vo
c
c
F
o
Q 5.14. The nominal interest rate is 20%. Inflation is 5%. What is the real interest rate?
n
po
l
an
(4
d
lch io
e
20
te
n
h
W
h
or
t
n
elc
(4 elc itio
a
.
t
di
ce
(4
ly
r
Inflation in Net Present Values

C
c
n
7. e (4 elc itio po
W
l
u
d
o
e
0
,
When it comes to inflation and net present value, there is a simple rule: Never mix apples and
J
W
(
elc iti
2
t
The most fundamental rule:

n
vo
oranges. The beauty of NPV is that every project’s cash flows are translated into the same units:
a
h
),
never mix apples and
ce
7
4
i
Ed
today’s dollars. Keep everything in the same units in the presence of inflation, so that this NPV
e
oranges. Nominal cash ows

n
01
tio po
l
Fi 7. I
(
W
advantage is not lost. When you use the NPV formula, always discount nominal cash flows with must be discounted with
4
,
e
W
nominal discount rates, and real (inflation-adjusted) cash flows with real (inflation-adjusted) nominal interest rates.
vo th
(
2
r
nc
a
o
discount rates.
7.
ce
)
(4
i
r
,C
F
a
01
Let’s return to our “apple” example. With 700% nominal interest rates and 100% inflation,
o
t
W
Our example discounted

n
the real interest rate is (1 + 700%)/(1 + 100%) – 1 = 300%. What is the value of a project that
e
7.
both in real and nominal

h
t
c
gives 12 apples next year, given that apples cost $1 each today and $2 each next year? terms.
a
.
t
1
e
7
4
or
There are two methods you can use:

0
c
F
on
01
I
(
u
y2
1. Discount the nominal cash flow of 12 apples next year ($2 · 12 = $24) with the nominal
p
h,
Discount nominal cash ows

.
01 nce
J
h
i
7
elc iti
interest rate. Thus, the 12 future apples are worth with nominal rates. Discount
F
elc
7.
t
1
,
l
real cash ows with real

)
v
Ju
Nominal Cash Flow $24

F
y 2 na
rates.
n
1
I
= $3
W
=
t
1 + Nominal Rate 1 + 700%

lch io
0
ra
7.
h
h
i
or
F
t
an vo
o
di
l
or , Ju e Fi
u
p
,C
2. Discount the real cash flows of 12 apples next year with the real interest rate. Thus, the
e
I
(
o W Ed Co , J
or
12 future apples are worth

u
n
17 ce
i
or
at
n)
eF
4t
l
d
u
e
io
l
I
(
E
J
r
n
ra
t
.
,
di
h
i
r
)
y
Co , Ju e F
po
t
l
)
4
v
d
u
h,
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
o
Real Cash Flow 12 Apples
01
ul
W
v
= 3 Apples
4
rp
=
h,
te
o
7.
1 + Real Rate 1 + 300%
a
I
(
,J
th
2
n
an vo
a
t
o
01
.
in today’s apples. Because an apple costs $1 today, the 12 apples next year are worth $3

di
n
i
lch ion or
C
y
F
o
today.
.I
na
ul
W
E
y2
Iv
p
,
e
o
Both the real and the nominal methods arrive at the same NPV result. The opportunity cost of
a
ce
J
r
at
th
i
y2
in
o
c
F
capital is that if you invest one apple today, you can quadruple your apple holdings by next year.
t
o
.
),
i
l
Iv
r
C
d
Thus, a 12-apple harvest next year is worth 3 apples to you today. The higher nominal interest
e
20
F
e
o
l
W
n
E
rates already reflect the fact that nominal cash flows next year are worth less than they are this
t
Ju
p
,
n
a
7.
ch
a
year. As simple as this may sound, I have seen corporations first work out the real value of their
or
th
ly
Fi
2
or
in
o
c
t
01
goods in the future, and then discount this with standard nominal interest rates. Just don’t!

di
l
n
Iv
Ju
,C
ly
4
e
F
o
na
W
(
oW E
y2
IMPORTANT
Ju
p
io
,
ce
or
)
h
i
2
• Discount nominal cash flows with nominal interest rates.
r
o
F
t
n
tio po
ul
4t
di
n
Iv
po ly
• Discount real cash flows with real interest rates.
lch io
e
0
e
o
a
e(
E
2
or
at
Ju
p
Either works. Never discount nominal cash flows with real interest rates, or vice-versa.
in
o
7 .
r
h
ly
i
or
C
c
c
t
o
n
01
i
l
n
v
Ju
r
4
d
e
rp
,
I
If you want to see this in algebra, the reason that the two methods come to the same result is
na
W
e(
2
Usually, it is best to work
o
that the inflation rate cancels out,
a

,
i
only with nominal quantities.
r
)
ly
i
Ed
iti
r
an vo
c
c
F
o
n
po
$24 12A 12A · (1 + 100%)
l
an
PV =
(4
d
= =
lch io
e
20
te
1 + 700% 1 + 300%
I (1 + 300%) · (1 + 100%)
n
h
W
h
or
t
n
elc
(4 elc itio
a
.
N R R · (1 + π)
t
di
ce
(4
ly
r
= = =
C
c
1 + rnominal 1 + rreal (1 + rreal ) · (1 + π)
n
7. e (4 elc itio po
W
l
u
d
o
e
0
,
where N is the nominal cash flow, r is the real cash flow, and π is the inflation rate. Most of the
J
W
(
elc iti
2
t
n
vo
.
time, it is easier to work in nominal quantities. Nominal interest rates are far more common
h
),
ce
7
4
i
Ed
r
e
than real interest rates, and you can simply use published inflation rates to adjust the future
n
01
tio po
l
Fi 7. I
(
W
4
price of goods to obtain future expected nominal cash flows.
,
e
W
vo th
(
2
r
nc
a
o
7.
ce
)
(4
i
r
,C
Q 5.15. If the real interest is 3% per annum and the inflation rate is 8% per annum, then what
F
a
01
o
t
W
is the present value of a $500,000 nominal payment next year?

n
e
7.
h
h
t
c
.
t
1
e
7
4
or
0
c
F
on
01
I
(
5.3 U.S. Treasuries and the Yield Curve

u
y2
p
h,

.
01 nce
J
h
i
7
elc iti
r
F
elc
7.
t
1
It is now time to talk in more detail about the most important financial market in the world
l
)
v
Ju
The simplest and most

e
F
today: the market for bonds issued by the U.S. government. These bonds are called Treasuries
y 2 na
1
I
important benchmark bonds

W
t
u
lch io
0
and are perhaps the simplest projects around. This is because, in theory, Treasuries cannot fail to
ra
nowadays are Treasuries.

7.
h
h
i
or
They have known and certain pay. They promise to pay U.S. dollars, and the U.S.-controlled Federal Reserve has the right to
F
t
an vo
o
di
l
payouts. print more U.S. dollars if it were ever to run out. Thus, there is no uncertainty about repayment
4
or , Ju e Fi
u
p
,C
e
for Treasuries. (In contrast, some European countries or U.S. states that borrow in currencies
(
o W Ed Co , J
or
u
n
that they cannot create may well not have the money to pay and therefore default.)
17 ce
i
or
at
n)
eF
The shorthand “Treasury” comes from the fact that the debt itself is issued by the U.S. Treasury
4t
l
U.S. Treasury bills, notes,

v
d
u
Department. There are three main types:

e
io
and bonds have different

l
I
(
E
J
r
n
ra
maturities. 1. Treasury bills (often abbreviated as T-bills) have maturities of up to one year.
t
.
,
di
h
i
r
)
y
Co , Ju e F
po
2. Treasury notes have maturities between one and ten years.

t
l
)
4
v
d
u
h,
0
a
3. Treasury bonds have maturities greater than ten years.

I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
5.3. U.S. Treasuries and the Yield Curve 87
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
The 30-year bond is often called the long bond. Together, the three are usually just called
o
n
01
ul
W
v
Treasuries. Conceptually, there is really no difference among them. All are really just obligations
4
rp
h,
te
o
7.
a
I
(
,J
issued by the U.S. Treasury. Indeed, there can be Treasury notes today that are due in 3 months—
th
2
n
an vo
a
t
o
01
.
such as a 9-year Treasury note that was issued 8 years and 9 months ago. This is really the same
di
n
i
lch ion or
C
y
F
o
obligation as a 3-month Treasury bill that was just issued. Thus, we shall be (very) casual with
.I
na
ul
W
E
y2
Iv
p
,
name distinctions.
e
o
h
a
ce
J
r
at
th
i
y2
in
o
In late 2015, the United States Federal Government owed over $18 trillion in Treasury
c
F
t
o
.
),
Magnitudes
i
l
Iv
r
C
obligations (on a GDP of about $17.5 trillion). With a population of 322 million, this debt
d
e
20
F
e
o
l
W
n
E
translated to over $55,000 per person. With 125 million households and 121 million full-time
t
Ju
p
,
n
a
7.
ch
a
workers, it represented about $150,000 per household or worker. (A worse problem, however, is
or
th
ly
Fi
2
or
in
o
c
t
01
that the United States has already promised benefits to future retirees that far exceed this number.)

di
l
n
Iv
Ju
,C
ly
4
e
F
But the United States also has assets. It owns more than $100 trillion in land, infrastructure,
o
na
W
(
oW E
y2
Ju
and mineral rights under the land.
p
io
,
ce
or
)
h
i
2
r
After Treasuries are sold by the government, they are then actively traded in what is one
o
F
t
n
tio po
The Treasuries market is

ul
4t
di
n
of the most important financial markets in the world today. It would not be uncommon for a
Iv
po ly
lch io
e
0
one of the most important

e
o
a
dedicated bond trader to buy $100 million of a Treasury note originally issued 10 years ago that e(
E
2
nancial markets in the

or
at
Ju
p
in
o
.
has 5 years remaining, and 10 seconds later sell $120 million of a three-year Treasury note issued world.
7
r
h
ly
i
or
C
c
c
6 years ago. Large buyers and sellers of Treasuries are easily found, and transaction costs are very
t
o
n
01
i
l
n
v
Ju
r
4
low. Trading volume is huge: Around 2015, it was about $500 billion per trading day. Therefore,
d
e
rp
,
I
na
W
e(
2
the annual trading volume in U.S. Treasuries—about 252 · $500 billion ≈ $130 trillion, where
o
a

,
i
252 is the approximate number of trading days per year—was about an order of magnitude
r
)
ly
i
Ed
iti
r
an vo
c
c
F
o
n
po
larger than the annual U.S. gross domestic product (GDP).
l
an
(4
d
lch io
e
20
te
Who owns them? About $6 trillion is owed to foreigners, with the Chinese (our largest
I
n
h
W
h
or
Who owns them?

t
n
elc
(4 elc itio
creditor) holding about $1.2 trillion of our bonds. about $12 trillion is owed to ourselves. The
a
.
t
di
ce
(4
ly
r
C
U.S. Federal Reserve estimated that domestically about 20% were held by individuals, 25%
c
n
7. e (4 elc itio po
W
l
u
by banks and mutual funds, 15% by public and private pension funds, 15% by state and local
d
o
e
0
,
W
governments, and 25% by other investors.
(
elc iti
2
t
n
vo
h
),
ce
Interest rates on Treasuries change every moment, depending on their maturity terms.
7
4
i
Ed
The yield curve shows the

e
01
Fortunately, you already know how to handle time-varying rates of return, so we can now put
tio po
l
Fi 7. I
(
W
annualized interest rate as a
4
,
your knowledge to the test. The principal tool for working with Treasury bonds is the yield curve
e
function of bond maturity.

h
W
vo th
(
2
r
nc
(or term structure of interest rates or just the term structure). It is a graphical representation,
a
o
7.
ce
)
(4
where the time to maturity is on the x-axis and the annualized interest rates are on the y-axis.
r
,C
F
a
01
o
t
W
There are also yield curves on non-Treasury bonds, but the Treasury yield curve is so prominent
u
n
e
7.
that unless clarified further, the yield curve should be assumed to mean investments in U.S.
h
h
t
c
Treasuries. (A more precise name would be the “U.S. Treasuries yield curve.”) This yield curve is
.
t
1
e
7
4
or
so important that most other debts in the financial markets, like mortgage rates or bank lending
0
c
F
on
01
I
(
u
y2
rates, are “benchmarked” relative to the Treasury yield curve. For example, if your firm wants to
p
h,

.
01 nce
J
h
i
issue a five-year bond, your creditors will want to compare your interest rate to that offered by
7
elc iti
r
F
elc
7.
t
1
equivalent Treasuries, and often will even describe your bond as offering “x basis points above
l
)
v
Ju
F
y 2 na
the equivalent Treasury.”

1
I
W
t
u
lch io
0
ra
7.
h
h
i
or
F
t
an vo
o
Q 5.16. What are the three types of Treasuries? How do they differ?
di
l
or , Ju e Fi
u
p
,C
e
I
(
o W Ed Co , J
or
u
n
17 ce
i
or
at
n)
eF
4t
l
d
u
e
io
l
I
(
E
J
r
n
ra
t
.
,
di
h
i
r
)
y
Co , Ju e F
po
t
l
)
4
v
d
u
h,
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
ce W di or Ju
l Fi 7. I (4 lch io
n or y2 an elc
7. (4 e l c t io p o n a vo t h , ) a t 01 ce h,
Iv th h, n) ra
y2
0 nc W Ed C , Ju e F 7. (4 Co
t e 1 e i
or
ly i Iv th
o W Ed Co , J
i r u l F i
7. ( 4
elc
h
tio po
r
na
n o E
n
88
e( 20
4t elch tion pora y 20 nan Ivo th E , Co ), J ate 17 ce ( We diti Annualized rate, r, in % Annualized rate, r, in %
h W
13
14
15
16
17
18
19
0
1
2
3
4
5
c
“inverted.”
E , C ) , t e 1 e e di rp ul Fi
n .I 4t lch
7 t
●
J
●
h
0
0
●
●●
W
●
●
u F . (4
●
●
di or i lch io or y2 an vo E ,C
l I n
●
●
po , ) at W d o
50
50
●
●
elc tio y 2 na vo th C , e 01 ce i
●
h h, n) ra W Ed Ju F 7. (4 elc tio
, t 01 nce or i
100
100
●
●
Ed Co J u eF 7. ( 4
elc iti
o po
r
ly
2
na Iv
o
th h,
C
n)
,
iti rp l y i n I v t h h , n a 0 n c W E d o
200
200
●
ly in Iv th h, n) ra
Co , Ju e F a t
Maturity (in months)

r p l i n
7. e (4 elc itio po
I t h n ra 2 n o E C , J
“Normal” (upward): May 2011
io h 01 ce W or
300
300
y2 an vo ,C ), te e di p ul
E t
Yield Curve Shapes
0 W o J F 7 ( l i o
“Inverted” (downward): December 1980

n) ora d u . 4 c o
r i
y2
●
●
,J te 17 ce
( i ly na Iv th h , n ) r at
rp u l F i . I 4 t
elc tio po
n r a n o E C , J
y n a v h h, ) t
20 c W d i o u
eF
or
e( el t i o
rp
o ly in
Ju e F 7. ( elc iti p o l y n I v
4t
h
ch
, n ) ra 20
Annualized rate, r, in % Annualized rate, r, in %
I 4t h on r a 2 an C ,J te 17
8.9
9.0
9.1
9.2
9.3
9.4
9.5
0
1
2
3
4
5
6
ly in v 0 c oW E o .
di r ul Fi
●
e p
0
0
n
●
t y
●
●●●
20 anc o W h E , Co ), J te F 17 e(
●
●
r u . 4 2
●
●
te 17 e di
t p l y i n I t h
lch ion or
a 0
an
50
50
●
●
e(
. 4 lch io or 2 an vo E ,C ), te 1 ce
●
Fi I t n a J 7
na h ) t ce W
100
100
vo ,C , e 01 di or Fi
●
●
n E e t p ul n .I
01 ce W d or J u F 7 . (4 lc io o y 2 a vo
7. i r at 01 n ce
(4 elch itio por ly 2 nan Ivo th E h, C n), e
200
200
W d Ju F 7. (4
Flat: January 2007
ce
●
01 or
Humped: June 1979

W ( e l i t i p l in Iv th

ce d or u F 7. 4 c o o y2 a o W
300
300

Iv n W d o Ju F . ( 4
elc
●
●
01 ce r
d u F 7 . ( 4 e l c
iti
o p o ly in
a I v t h h,
shape. In 2016, the yield curve was normal, but not as steep as that in May 2011. A downward slope is sometimes called
r 2
Exhibit 5.1: Various Historical Yield-Curve Shapes. The upward slope is so common that it is considered the “normal”
Time-Varying Rates of Return and the Yield Curve
oW E
(4 elch itio por ly 2 inan Ivo th E h, C n), rate 017 nce e di
th n ) at 0 c e W d o Ju F . ( 4 l c tio
,C 1 r po ly in Iv th h,
e
Ed
i
or , Ju e Fi
p l
7. (4 elc itio
t h n r 2 a n o E C
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
Exhibit 5.1 shows some historical yield curves. They are commonly classified into four basic
o
n
01
Yield curves are often, but
ul
W
v
shapes:
4
rp
h,
te
o
not always, upward-sloping.
7.
a
I
(
,J
th
2
1. Flat: There is little or no difference between annualized short-term and long-term rates. A
n
an vo
a
t
o
01
.
di
n
i
flat yield curve is basically the scenario that was the subject of the previous chapter. It
lch ion or
C
y
F
o
.I
na
n
means you can simplify (1 + r0,t ) ≈ (1 + r)t .
ul
W
E
y2
Iv
p
,
e
o
h
a
ce
J
2. Upward-sloping (“normal”): Short-term rates are lower than long-term rates. This is the
at
th
i
y2
in
o
c
F
t
o
.
most common shape. It means that longer-term interest rates are higher than shorter-term
),
i
l
Iv
r
C
d
e
20
F
e
interest rates. Since 1934, the steepest yield curve (the biggest difference between the
o
l
W
n
E
t
Ju
p
,
n
long-term and the short-term Treasury rates) occurred in October 1992, when the long-term
a
7.
ch
a
or
th
ly
Fi
2
or
interest rate was 7.3% and the short-term interest rate was 2.9%—just as the economy
in
o
c
t
01

di
l
n
pulled out of the recession of 1991. As of 2016, the yield curve has been upward-sloping
Iv
Ju
,C
ly
4
e
F
o
na
W
since the Great Recession. (
oW E
y2
Ju
p
io
,
ce
3. Downward-sloping (“inverted”): Short-term rates are higher than long-term rates.
or
)
h
i
2
r
o
F
t
n
tio po
ul
4t
di
n
4. Humped: Short-term rates and long-term rates are lower than medium-term rates.
Iv
po ly
lch io
e
0
e
o
a
e(
E
2
or
Inverted and humped yield curves are relatively rare.

at
Ju
p
in
o
7 .
r
h
ly
i
or
C
c
c
t
o
n
01
i
l
n
Macroeconomic Implications of Different Yield Curve Shapes v

Ju
r
4
d
e
rp
,
I
na
W
e(
2
o
Economists and pundits have long wondered what they can learn from the shape of the yield curve about the future of the
a

,
i
r
)
ly
i
Ed
iti
r
economy. It appears that the yield curve shape is a useful—though unreliable and noisy—signal of where the economy is
an vo
c
c
F
o
n
po
l
an
heading. Steep yield curves often signal emergence from a recession, as you will see in Exhibit 5.4 on Page 96. Inverted
(4
d
lch io
e
20
te
n
h
W
yield curves often signal an impending recession. But can’t the Federal Reserve Bank control the yield curve and thereby
h
or
t
n
elc
(4 elc itio
a
.
t
control the economy? It is true that the Fed can influence the yield curve—and since 2008, it has worked on influencing it
di
ce
(4
ly
r
C
c
like never before. But ultimately the Fed does not control it—instead, it is the broader demand and supply for savings and
n
7. e (4 elc itio po
W
l
u
d
o
e
0
credit in the economy that determines it. Economic research has shown that the Fed typically has a good deal of influence
,
W
(
elc iti
2
t
on the short end of the Treasury curve—by expanding and contracting the supply of money and short-term loans in the
n
vo
h
),
ce
7
economy—but not much influence on the long end of the Treasury curve, especially in the long-run. And even in the
4
i
Ed
r
e
01
tio po
l
Fi 7. I
financial crisis of 2008, the Fed’s influence on the short end was ultimately limited, too—the nominal rate already stood at
(
W
4
,
0% and there was little the Fed could do to drop it further. (Large negative nominal rates are not possible.) In fact, by
e
W
vo th
(
2
r
nc
flooding the economy with cheap money, the Fed was trying to push banks to lend and people to spend money—but people
a
o
7.
ce
)
(4
i
r
instead just deposited the cash right back into the banks!
,C
F
a
01
o
t
W
u
n
e
7.
h
h
t
c
.
t
1
If you want to undertake your own research, you can find historical interest rates at the St.
e
7
4
or
Common data sources for

0
c
F
on
Louis Federal Reserve Bank at http://research.stlouisfed.org/fred. There are also the Treasury
01
I
(
interest rates.
u
y2
p
h,
Management Pages at http://www.tmpages.com/. Or you can look at SmartMoney.com for

.
01 nce
J
h
i
7
elc iti
historical yield curves. PiperJaffray.com has the current yield curve—as do many other financial
F
elc
7.
t
1
,
l
sites and newspapers. Finance.yahoo.com/bonds provides not only the Treasury yield curve but
)
v
Ju
F
y 2 na
1
I
also yield curves for many other types of bonds.

W
t
u
lch io
0
ra
7.
h
h
i
or
F
t
an vo
o
An Example: The Yield Curve on December 31, 2015

t
di
l
or , Ju e Fi
u
p
,C
e
I
(
o W Ed Co , J
or
Let’s focus on working with one particular yield curve. Exhibit 5.2 shows the Treasury yields
u
n
We will analyze the

17 ce
on December 31, 2015. This yield curve had the most common upward slope. The curve tells
i
or
Treasury yield curve at the

at
n)
eF
you that if you had bought a 3-month Treasury at the end of the day on December 31, 2015,
4t
l
end of December 2015.

v
d
u
e
io
your annualized interest rate would have been 0.16% per annum. (A $100 investment would
n
l
I
(
E
J
r
1
n
turn into $100 · (1 + 0.16%) /4 ≈ $100 · 1.0003998 ≈ $100.04 on March 31, 2016.) If you had
ra
t
.
,
di
h
i
r
)
y
Co , Ju e F
po
t
l
)
4
v
d
u
h,
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
iti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
Ed
eF
vo
elc
ul
di
an
h,
(4
C
tio
20
ch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
l
h
Fi
r
C
n)
ly
(4
d
e
c
o
t
W
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
bought a thiry-year bond, your annualized interest rate would have been 3.01% per annum, or
o
n
01
ul
W
(
v
$243.43 on December 2045 for a $100 investment.
4
rp
h,
te
o
7.
a
I
e
(
,J
th
Sometimes it is necessary to determine an interest rate for a bond that is not listed. This
2
n
o
nc
a
t
o
You can interpolate
01
.
di
is usually done by interpolation. For example, if you had wanted to find the yield for a 4-
v
i
lch ion or
C
y
annualized interest rates on
o
I
na
n
year bond, a reasonable guess would have been an interest rate halfway between the 3-year
ul
W
the yield curve.
E
y2
Iv
p
,
e
o
.
bond and the 5-year bond. In December 2015, this would have been an annualized yield of
a
ce
7
J
r
at
th
i
y2
in
o
(1.31% + 1.76%)/2 ≈ 1.54%. (This is not exact, as you can guess by noticing that the yield curve
c
F
t
o
1
.
),
i
l
an
Iv
r
C
looks more concave.)
d
e
n) r, in % rat 20
F
e
o
l
W
n
E
t
Ju
p
,
in
a
7.
ch
a
or
th
ly
2
or
in
o
c
eF
t
01

di
l
n
Iv
Ju
,C
ly
4
e
F
o
na
W
3.0 (
oW E
2
Ju
p
io
,
ce
or
)
h
y
2
o
F
t
●
n
tio po
ul
4t
di
2.5
n
Iv
po ly
lch io
e
0
e
o
a
,J
e(
E
2
or
●
at
Ju
p
in
o
.
●
2.0
7
r
h
ly
i
r
C
rate,
c
c
t
o
01
o
i
l
n
v
Ju
r
4
d
e
rp
,
I
na
W
1.5
e(
2
4t elch tion pora y 20 nan Ivo th E Annualized
o
a
Co

,
i
r
)
ly
i
Ed
iti
r
an vo
c
c
F
o
n
po
1.0
●
l
an
(4
d
o
e
20
h,
te
n
W
h
i
or
t
n
elc
(4 elc itio
●
a
.
t
0.5
ce
(4
ly
●
d
r
C
c
n
7. e (4 elc itio po
W
l
u
d
o
e
0
,
●●
h
0.0
J
W
(
elc iti
2
t
n
vo
h
),
ce
0 50 100 150 200 250 300 350
7
l
4
i
Ed
r
e
01
tio po
l
Fi 7. I
(
W
4
,
e
W
vo th
(
2
r
nc
a
o
7.
ce
)
4
i
r
,C
←− Months Years −→
F
a
01
o
l
e(
t
W
1 3 6 1 2 3 5 7 10 20 30
u
n
e
7.
h
h
12/31/2015 0.14 0.16 0.49 0.65 1.06 1.31 1.76 2.09 2.27 2.67 3.01
c
.
t
1
e
7
4
or
0
c
F
Exhibit 5.2: The Treasury Yield Curve on December 31, 2015. These rates are annualized yields to maturity (internal rates
on
01
I
(
u
y2
p
h,
of return) calculated from Treasury prices. If they were truly Treasury zero-bonds, they would just be the standard discount
.
01 nce
J
h
i
17
elc iti
rates computed from the final payment and today’s price, but we ignore the details here. Such yield curves can be found on
F
elc
7.
t
,
l
many websites. The yield curve changes every day—although day-to-day changes are usually small. Our example works
)
v
Ju
F
y 2 na
1
I
primarily with this particular yield curve. Source: Federal Reserve, www.federalreserve.gov/releases/h15/data.htm.
W
t
u
lch io
0
ra
7.
h
h
i
or
F
t
an vo
o
di
l
or , Ju e Fi
u
p
,C
te
I
(
o W Ed Co , J
or
u
n
As notation for the annualized horizon-dependent interest rates, we continue using our
17 ce
The December 2015 yield

i
or
at
earlier method. We call the two-year annualized interest rate r2 (here, 1.06%), the three-year
n)
eF
curve was upward-sloping:

4t
l
Annualized interest rates

u
e
io
Deeper: There are some small inaccuracies in my description of yield curve computations. My main
n
l
I
(
were higher for longer

E
,J
r
simplification is that U.S. yield curves are based on semi-annually-compounded coupon bonds in real life,
n
ra
t
maturities.
.
whereas our textbook pretends that the yield is quoted on a zero bond. In corporate finance, the yield
di
h
i
r
)
y
Co , Ju e F
difference between annual compounding and semi-annual compounding is almost always inconsequential.
po
t
l
However, if you want to become a fixed-income trader, you cannot take this approximation literally. Consult
)
4
v
d
u
h,
0
a
a dedicated fixed-income text instead.

I
e(
J
r
W
y2
n
c
p
o
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
annualized interest rate r3 (here, 1.31%), and so on. It is always these overlined-subscript yields
o
n
01
ul
W
v
that are graphed in yield curves. Let’s work with this particular yield curve, assuming it is based
4
rp
h,
te
o
7.
a
I
(
,J
exclusively on zero-bonds, so you don’t have to worry about interim payments.
th
2
n
an vo
a
t
o
01
.
di
n
i
Holding rates of return First, let’s figure out how much money you will have at maturity. That
lch ion or
C
y
F
o
Computing the holding rate
.I
na
ul
W
is, how much does an investment of $500,000 in U.S. two-year notes (i.e., a loan to the
E
y2
Iv
of return for 2-year and
p
,
e
o
h
a
U.S. government of $500,000) on December 31, 2015, return on December 31, 2017? Use
ce
30-year Treasuries.
J
r
at
th
i
y2
in
o
c
F
t
o
the data in Exhibit 5.2. Because the yield curve prints annualized rates of return, the total
.
),
i
l
Iv
r
C
d
e
20
two-year holding rate of return (as in Formula 5.1) is the twice compounded annualized
F
e
o
l
W
n
E
t
rate of return,
Ju
p
,
n
a
7.
ä Annualizing,
ch
a
or
th
ly
Fi
2
or
Formula 5.1, Pg.78.
in
o
c
t
01
r0,2 = 1.0106 · 1.0106 – 1 ≈ 2.13%

di
l
n
Iv
Ju
,C
ly
4
e
F
o
na
W
r0,2 = (1 + r2 ) · (1 + r2 ) – 1 (
oW E
y2
Ju
p
io
,
ce
or
)
h
i
2
r
so your $500,000 will turn into

o
F
t
n
tio po
ul
4t
di
n
Iv
po ly
lch io
e
0
e
o
a
C2 ≈ (1 + 2.13%) · $500,000 ≈ $510,656 e(
E
2
or
at
Ju
p
in
o
7 .
r
h
ly
i
or
C2 = (1 + r0,2 ) · C0
C
c
c
t
o
n
01
i
l
n
v
Ju
r
4
d
e
rp
,
on December 31, 2017. (In the real world, you might have to pay a commission to arrange
I
na
W
e(
2
this transaction, so you would end up with a little less.)
o
a

,
i
r
)
ly
i
Ed
iti
r
an vo
What if you invest $500,000 into 30-year Treasuries? Your 30-year holding rate of return
c
c
F
o
n
po
l
an
is
(4
d
lch io
e
20
te
n
h
W
h
or
t
n
elc
(4 elc itio
a
.
r0,30 = 1.030130 – 1
t
≈ 2.434 – 1 ≈ 143.4%
di
ce
(4
ly
r
C
c
n
7. e (4 elc itio po
30
W
l
r0,30 = (1 + r30 ) –1
u
d
o
e
0
,
W
(
elc iti
2
t
Thus, an investment of C0 = $500,000 in December 2015 turns into a return of C30 ≈ $1.2
vo
h
),
ce
7
4
i
Ed
million by December 2045.

e
01
tio po
l
Fi 7. I
(
W
4
Forward rates of return Second, let’s figure out what the yield curve in December 2015 implied
,
e
Let's work out one forward
W
vo th
(
2
r
about the 1-year interest rate from December 2016 to December 2017. This would be best
nc
rate implied by the

a
o
7.
ce
named r1,2 . It is an interest rate that begins in one year and ends in two years. As already
)
(4
i
r
December 2015 yield curve.

,C
F
a
01
o
mentioned, this is called the forward rate.
t
W
u
n
e
7.
The 1-year annualized interest rate is r1 = 0.65%. The two-year annualized rate of return
h
h
t
c
is r2 = 1.06%. You already know that you can work out the two holding rates of return,
.
t
1
e
7
4
or
r0,1 = 0.65% and r0,2 = (1 + r2 )2 – 1 ≈ 1.0213%. You only need to use the compounding
0
c
F
on
01
I
(
u
y2
formula to determine r1,2 :

p
h,

.
01 nce
J
h
i
7
elc iti
r
F
elc
7.
t
1
,
l
(1 + 2.13%) = (1 + 0.65%) · (1 + r1,2 ) ⇒ r1,2 ≈ 1.47%

)
v
Ju
F
y 2 na
1
I
W
t
(1 + r0,2 ) = (1 + r0,1 ) · (1 + r1,2 )

lch io
0
ra
7.
h
h
i
or
F
t
an vo
o
Note that this forward rate r1,2 is higher than both r1 and r2 from which you computed it.
t
di
l
or , Ju e Fi
u
p
,C
e
Exhibit 5.3 summarizes our two-year calculations, and extends them by another year. (This
I
(
o W Ed Co , J
or
Is the proliferation of
n
helps you to check your results in an exercise below.) One question you should ask yourself is
17 ce
subscripts torture or
i
or
at
n)
eF
whether I use so many subscripts in the notation just because I enjoy torturing you. The answer necessity?
4t
l
d
u
is an emphatic no: The subscripts are there for good reason. When you look at Exhibit 5.3, for
e
io
l
I
(
E
J
r
example, you have to distinguish between the following:

n
ra
t
.
,
di
h
i
• the three holding rates of return, r0,t (0.65%, 2.13%, and 3.98%)
)
y
Co , Ju e F
po
t
l
)
4
v
• the three annualized rates of return, rt (0.65%, 1.06%, and 1.31%)

u
h,
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
o
n
01
ul
W
v
4
rp
h,
te
o
7.
a
I
(
,J
th
2
n
an vo
a
t
o
01
.
di
n
i
lch ion or
C
y
Rates of Return
o
.I
na
ul
W
E
Maturity Total Holding Annualized Compounded Rates
y2
Iv
p
,
e
o
h
a
ce
J
r
at
th
i
= (1 + 0.65%)1
y2
1 Year (1 + 0.65%) = (1 + 0.65%)
in
o
c
F
t
o
.
),
i
= (1 + r1 )1
l
Iv
(1 + r0,1 ) = (1 + r0,1 )
r
C
d
e
20
F
e
o
l
W
n
E
t
Ju
p
,
n
≈ (1 + 1.06%)2
a
7.
2 Years (1 + 2.13%) ≈ (1 + 0.65%) · (1 + 1.47%)
ch
a
or
th
ly
Fi
2
or
in
= (1 + r2 )2
o
c
(1 + r0,2 ) = (1 + r0,1 ) · (1 + r1,2 )
t
01

di
l
n
Iv
Ju
,C
ly
4
e
F
o
na
≈ (1 + 1.31%)3
W
3 Years (1 + 3.98%) (
≈ (1 + 0.65%) · (1 + 1.47%) · (1 + 1.81%)
oW E
y2
Ju
p
io
,
= (1 + r3 )3
ce
(1 + r0,3 ) = (1 + r0,1 ) · (1 + r1,2 ) · (1 + r2,3 )
or
)
h
i
2
r
o
F
t
n
tio po
ul
4t
di
n
Iv
po ly
lch io
e
0
e
o
a
e(
E
2
or
at
Ju
p
in
o
.
Exhibit 5.3: Relation between Holding Returns, Annualized Returns, and Year-by-Year Returns on December 31, 2015, by
7
r
h
ly
i
or
C
c
c
t
o
n
01
Formula. The individually compounded rates are the future interest rates. They are implied by the annualized rates quoted
i
l
n
v
Ju
r
4
d
e
rp
,
in the middle column. The text worked out the two-year case. You will work out the three-year case in Question 5.17.
I
na
W
e(
2
o
a

,
i
r
)
ly
i
Ed
iti
r
an vo
c
c
F
o
n
po
l
an
(4
d
lch io
e
20
te
n
h
W
h
or
t
n
elc
(4 elc itio
• the three individual annual rates of return rt–1,t (0.65%, 1.47%, and 1.81%), where the
a
.
t
di
ce
(4
ly
r
C
second and third forward rates begin at different moments in the future.
c
n
7. e (4 elc itio po
W
l
u
d
o
e
0
In real life, you have not just three yearly Treasuries, but many Treasuries between 1 day and 30
,
W
(
elc iti
2
t
years. Anyone dealing with Treasuries (or CDs or any other fixed-income investment) that can
n
vo
h
),
ce
7
have different maturities or start in the future must be prepared to suffer double subscripts.
4
i
Ed
r
e
01
tio po
l
Fi 7. I
(
W
If Treasuries offer different annualized rates of return over different horizons, do corporate
4
Yes, corporate projects
,
e
projects have to do so, too? Almost surely yes. If nothing else, they compete with Treasury
W
vo th
(
2
have double subscripts, too!

r
nc
a
o
bonds for investors’ money. And just like Treasury bonds, many corporate projects do not begin
7.
ce
)
(4
i
r
,C
immediately, but may take a year or more to prepare. Such project rates of return are essentially
F
a
01
o
t
W
forward rates of return. Double subscripts—yikes, but sometimes there is no way out of painful
n
e
7.
h
h
notation in the real world!

t
c
.
t
1
e
7
4
or
0
c
F
on
01
I
(
u
y2
Q 5.17. Compute the three-year holding rate of return on December 31, 2015. Then, using the
p
h,

.
01 nce
J
h
two-year holding rate of return on December 31, 2015, and your calculated three-year holding
i
7
elc iti
r
F
elc
7.
t
1
rate of return, compute the forward interest rate for a 1-year investment beginning on December
l
)
v
Ju
F
31, 2017, and ending on December 31, 2018. Are these the numbers in Exhibit 5.3?
y 2 na
1
I
W
t
u
lch io
0
ra
7.
Q 5.18. Repeat the calculation with the five-year annualized rate of return of 1.76%. That is,
h
h
i
or
F
t
an vo
what is the five-year holding rate of return, and how can you compute the annualized forward
o
di
l
or , Ju e Fi
u
p
,C
interest rate for a two-year investment beginning on December 31, 2018, and ending on December
e
I
(
o W Ed Co , J
or
31, 2020?
n
17 ce
i
or
at
n)
eF
4t
l
d
u
e
io
l
I
(
E
J
r
n
ra
t
.
,
di
h
i
r
)
y
Co , Ju e F
po
t
l
)
4
v
d
u
h,
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
Bond Payoffs and Your Investment Horizon
y
d
o
n
01
ul
W
v
4
rp
h,
te
o
7.
Should there be a link between your personal investment horizon and the kinds of bonds you
a
I
(
,J
th
Your investment horizon has
2
n
an vo
may be holding? Let’s say that you want to buy a three-year zero-coupon bond because it offers
a
t
o
01
.
no link to the time patterns

di
n
i
lch ion or
C
y
1.31%, which is more than the 0.65% that a 1-year zero-coupon bond offers—but you also want
F
of bond payoffs you invest
o
.I
na
ul
W
E
y2
to consume in one year. Can you still buy the longer-term bond? There is good news and bad in. You can always sell
Iv
p
,
e
o
h
a
ce
news. The good news is that the answer is yes: There is no link whatsoever between your desire long-term bonds to get
J
r
at
th
i
y2
in
o
c
F
t
o
money quickly, if need be.
.
to get your money back and consume, and the point in time when your three-year bond pays off.
),
i
l
Iv
r
C
d
e
20
F
e
You can always buy a three-year bond today, and sell it before maturity, such as next year when it
o
l
W
n
E
t
Ju
p
will have become a two-year bond. The bad news is that in our perfect and certain market, this
,
n
a
7.
ch
a
or
th
ly
Fi
investment strategy will still only get you the 0.65% that the 1-year bond offers. If you buy $100
2
or
in
o
c
t
01
of the three-year bond for P = $100/1.01313 ≈ $96.17 today, next year it will be a two-year

di
l
n
Iv
Ju
,C
ly
4
e
F
o
na
bond with an interest rate of 1.47% in the first year and 1.81% in the second year (both worked
W
(
oW E
y2
Ju
p
io
out in Exhibit 5.3). You can sell this bond next year for a price of
,
ce
or
)
h
i
2
r
o
F
t
n
tio po
ul
$100 $100 $100
4t
di
n
Iv
po ly
≈ $96.80
lch io
e
0
= =
e
o
1 + r1,3 (1 + r1,2 ) · (1 + r2,3 ) 1.0147 · 1.0181
a
e(
E
2
or
at
Ju
p
in
o
7 .
Your 1-year holding rate of return would therefore be only ($96.80 – $96.17)/ $96.17 ≈ 0.65%—
r
h
ly
i
or
C
c
c
t
o
n
01
t
the same rate of return you would have received if you had bought a 1-year bond to begin with.
i
l
n
v
Ju
r
4
d
e
rp
,
There is no free lunch here.

I
na
W
e(
2
o
a

,
i
r
)
ly
i
Ed
iti
r
The Effect of Interest Rate Changes on Short-Term and Long-Term Bonds
an vo
c
c
F
o
n
po
l
an
(4
d
lch io
e
20
te
n
h
Are long-term bonds riskier than short-term bonds? Of course, recall that repayment is no
W
h
or
Treasuries pay what they

t
n
elc
(4 elc itio
a
.
less certain with long-term Treasury bonds than short-term Treasury notes. (This would be an
t
di
ce
promise. They have no
(4
ly
r
C
c
issue of concern if you were to evaluate corporate projects that can go bankrupt. Long-term
n
default risk. They do have

7. e (4 elc itio po
W
l
u
4
corporate bonds are often riskier than short-term corporate bonds—most firms are unlikely to
d
o
the risk of interim interest
e
0
,
W
(
elc iti
go bankrupt this week, but more likely to go bankrupt over a multidecade time horizon.) So, for rate changes.
2
t
n
vo
h
),
ce
Treasury bonds, as long as Congress does not go crazy, there should be no uncertainty as far as
7
4
i
Ed
r
e
01
payment uncertainty is concerned. But there may still be some interim risk of a different kind;
tio po
l
Fi 7. I
(
W
4
,
and even though we have not yet fully covered it, you can still intuitively figure out why this
e
W
vo th
(
2
r
nc
is so. Ask yourself how economy-wide bond prices (interest rates) can change in the interim
a
o
7.
ce
)
(4
(before maturity). What are the effects of sudden interest rate changes before maturity on bond
r
,C
F
a
01
o
t
W
values? It turns out that an equal-sized interest rate movement can be much more dramatic for
u
n
e
7.
long-term bonds than for short-term bonds. Let me try to illustrate why.
h
h
t
c
.
t
1
e
The 30-year bond: Work out the value of a $1,000 30-year zero-bond at the 3.01% interest rate
7
4
or
First, the effect of a 10 bp

0
c
F
on
01
prevailing. It costs $1,000/1.030130 ≈ $410.79. You already know that when prevailing
I
(
yield change on the price of

u
y2
p
h,

.
interest rates go up, the prices of outstanding bonds drop and you will lose money. For
01 nce
a 30-year bond.
J
h
i
7
elc iti
r
F
example, if interest rates increase by 10 basis points to 3.11%, the bond value decreases to
elc
7.
t
1
,
l
$1,000/1.031130 ≈ $399.00. If interest rates decrease by 10 basis points to 2.91%, the

v
Ju
F
y 2 na
1
I
bond value increases to $1,000/1.029130 ≈ $422.93. Thus, the effect of a 10-basis-point

t
u
lch io
0
ra
7.
change in the prevailing 30-year yield induces an immediate percent change (an instant
h
i
or
F
t
an vo
o
rate of return) in the value V of your bond of

di
l
or , Ju e Fi
u
p
,C
e
I
(
o W Ed Co , J
or
u
n
V(r30 = 3.11%) – V(r30 = 3.01%)

17 ce
$399.00 – $410.79
i
or
Up 10 bp: r ≈ ≈ –2.87%
at
n)
=
eF
V(r30 = 3.01%) $410.79

4t
l
d
u
V(r30 = 2.91%) – V(r30 = 3.01%)

io
$422.93 – $410.79
n
l
I
(
E
J
r
Down 10 bp: r =
t
≈ ≈ +2.96%
n
V(r30 = 3.01%) $410.79

ra
t
.
,
di
h
i
r
)
y
Co , Ju e F
po
So for every $1 million you invest in 30-year bonds, you expose yourself to about $30,000
t
l
)
4
v
d
u
h,
in instant risk for every 10-basis-point yield change in the economy.

a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
The 1-year note: To keep the example identical, let’s now assume that the 1-year note also has
o
n
01
Second, the effect of a
ul
W
v
an interest rate of 3.01% and consider the same 10-basis-point change in the prevailing
4
rp
h,
te
o
10 bp point change on the
7.
a
I
(
,J
interest rate. In this case, the equivalent computations for the value of a 1-year note are
th
2
price of a 1-year note.
n
an vo
a
t
o
01
.
$970.78 at 3.01%, $971.72 at 2.91%, and $969.84 at 3.11%. Therefore, the equivalent
di
n
i
lch ion or
C
y
F
o
instant rates of return are
.I
na
ul
W
E
y2
Iv
p
,
e
o
h
a
ce
J
r
at
th
i
y2
in
o
V(r1 = 3.11%) – V(r1 = 3.01%)
c
F
$971.72 – $970.78
t
o
.
),
Up 10 bp: r ≈ ≈ –0.097%
i
l
=
Iv
r
C
d
V(r1 = 3.01%) $970.78
e
20
F
e
o
l
W
n
E
t
Ju
V(r1 = 3.11%) – V(r1 = 3.01%) $969.84 – $970.78
p
,
n
a
7.
ch
a
Down 10 bp: r = ≈ ≈ +0.097%
or
th
ly
Fi
V(r1 = 3.01%) $970.78
2
or
in
o
c
t
01

di
l
n
Iv
Ju
,C
ly
4
e
For every $1 million you invest in 1-year notes, you expose yourself to about $1,000 risk
F
o
na
W
(
oW E
y2
Ju
in instant risk for every 10-basis-point yield change in the economy.
p
io
,
ce
or
)
h
i
2
r
It follows that the value effect of an equal-sized change in prevailing interest rates is more
o
F
t
n
tio po
ul
4t
An equal interest rate move
di
n
Iv
severe for longer-term bonds. In turn, it follows that if the bond is due tomorrow, interest rate
po ly
lch io
e
0
e
affects longer-term bonds
o
a
e(
changes can usually wreak very little havoc. You will be able to reinvest tomorrow at whatever
E
2
or
at
more strongly.
Ju
p
in
o
.
the new rate will be. A long-term bond, on the other hand, may lose (or gain) a lot of value.
7
r
h
ly
i
or
C
c
c
t
o
n
01
In sum, you should always remember that Treasury bonds are risk-free in the sense that
i
l
n
v
Ju
r
Again, in the interim,
d
e
rp
,
they cannot default (fail to return the promised payments), but they are risky in the sense that
I
na
W
T-bonds are not risk-free!
e(
2
interim interest changes can alter their values. Only the most short-term Treasury bills (say,
o
a

,
i
r
)
ly
i
Ed
iti
r
due overnight) can truly be considered risk-free—virtually everything else suffers interest-rate
an vo
c
c
F
o
n
po
l
change risk.
an
(4
d
lch io
e
20
te
n
h
W
h
or
t
IMPORTANT
elc
(4 elc itio
a
.
t
di
ce
(4
ly
Though “fixed income,” even Treasuries do not guarantee a “fixed rate of return” over horizons
r
C
c
n
7. e (4 elc itio po
W
shorter than their maturities. Day to day, long-term Treasury bonds are generally riskier in-
l
u
d
o
e
0
,
vestments than short-term Treasury bills, because interest-rate changes have more impact on
J
W
(
elc iti
2
t
n
vo
them.
.
h
),
ce
7
4
i
Ed
r
e
01
tio po
l
Fi 7. I
(
W
4
,
e
Confession time: I have pulled two cheap tricks on you. First, in the real world, what if
W
vo th
(
2
r
For illustration, I have

nc
a
o
7.
short-term, economy-wide interest rates typically experienced yield shifts of plus or minus 100
ce
ignored volatility of changes
)
(4
i
r
,C
basis points, while long-term, economy-wide interest rates never budged? If this were true,
F
a
and earned interest.

01
o
t
W
long-term bonds could even be safer than short-term bonds. However, the empirical evidence
n
e
7.
h
h
from 1990 to 2016 suggests that day-to-day changes of both were of similar magnitude—about
t
c
.
t
1
e
plus or minus 5 basis points a day. (One-month yields changed by about 7 basis points a day.)
7
4
or
0
c
F
on
01
Second, I ignored that between today and tomorrow, you would also earn 1 day of interest. On
I
(
u
y2
p
h,
a $1,000,000 investment in 1-years, this would be about $25; in 30-years, about $120. Thus
.
01 nce
J
h
i
7
elc iti
about $100 should be added to the long-term bond investment strategy—but $100 on a $30,000
F
elc
7.
t
1
,
l
risk exposure was small enough to keep ignorance bliss.

v
Ju
F
y 2 na
1
I
W
t
u
lch io
0
ra
7.
h
h
i
or
Q 5.19. A ten-year and a 1-year zero-bond both offer an interest rate of 8% per annum.
F
t
an vo
o
di
l
or , Ju e Fi
u
p
,C
e
1. How does an increase of 1 basis point in the prevailing interest rate change the value of
I
(
o W Ed Co , J
or
the 1-year bond? (Use 5 decimals in your calculation.)

n
17 ce
i
or
at
n)
2. How does an increase of 1 basis point in the prevailing interest rate change the value of
eF
4t
l
the ten-year bond?

v
d
u
e
io
l
I
(
E
J
r
3. What is the ratio of the value change over the interest change? (In calculus, this would be
n
ra
t
.
,
di
called the derivative of the value with respect to interest rate changes.) Which derivative
h
i
r
)
y
Co , Ju e F
po
is larger?
l
)
4
v
d
u
h,
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
5.4. Why Does the Yield Curve Usually Slope Up? 95
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
A Tragic Error: “Paper Losses”?!
y
d
o
n
01
ul
W
v
4
rp
h,
te
o
7.
If you really need cash from a bond investment in 20 years, doesn’t a prevailing interest rate
a
I
(
,J
th
“Only” a paper loss: A
2
n
an vo
increase cause only an interim paper loss? This is a capital logical error many investors commit.
a
t
o
01
.
cardinal error!

di
n
i
lch ion or
C
y
Say that a 10-basis-point increase happened overnight, and you had invested $1 million yesterday.
o
.I
na
ul
W
E
y2
You would have lost $30,000 of your net worth in 1 day! Put differently, waiting 1 day would
Iv
p
,
e
o
ä Tax treatment of realized and
a
ce
have saved you $30,000 or allowed you to buy the same item for $30,000 less. Paper money
J
r
at
th
i
unrealized capital gains,
y2
in
o
c
F
t
o
Sect. 11.4, Pg.257.
.
is actual wealth. Thinking paper losses are any different from actual losses is a common but
),
i
l
Iv
r
C
d
e
20
F
e
capital error. (The only exception to this rule is that realized gains and losses have different
o
l
W
n
E
t
Ju
p
tax implications than unrealized gains and losses.) Avoiding this conceptual mistake is more
,
n
a
7.
ch
a
or
th
ly
Fi
important than learning any formulas in this book.
2
or
in
o
c
t
01

di
l
n
Iv
Ju
,C
ly
4
e
F
IMPORTANT
o
na
W
(
oW E
y2
Ju
“Paper losses” are no less real than realized losses.
p
io
,
ce
or
)
h
i
2
r
o
F
t
n
tio po
ul
4t
di
n
Iv
po ly
lch io
e
0
e
o
a
e(
E
2
or
at
Ju
p
in
5.4 Why Does the Yield Curve Usually Slope Up?
o
7 .
r
h
ly
i
or
C
c
c
t
o
n
01
i
l
n
v
Ju
r
Aren’t you already wondering why the yield curve is not usually flat? Take our example yield
d
e
rp
,
But why? Why? Why?

I
na
W
e(
curve from December 2015. Why did the 30-year Treasury bonds in December 2015 pay 3.01%
2
o
a

,
i
per year, while the one-month Treasury bills paid only 0.14% per year? And why is the upward
r
)
ly
i
Ed
iti
r
an vo
c
c
F
o
n
slope the most common shape?

po
l
an
(4
d
lch io
e
20
te
First, let’s look at the historical data. We cannot easily visualize the entire historical yield
I
n
h
W
h
Historical Yield Curves

or
t
curve in a two-dimensional graph, but we can plot the historical yields on the short-term 3-month
elc
(4 elc itio
a
.
t
di
ce
(4
ly
r
Treasury and, say, the 20-year Treasury. Exhibit 5.4 shows that the spread between them ranged
C
c
n
7. e (4 elc itio po
W
l
from negative (though usually only briefly) to about 300 bp per year. (Humped shapes are rare,
u
d
o
e
0
,
so if the long-term rate is below the short-term rate, the yield curve is most likely inverted.) The
W
(
elc iti
2
t
n
vo
.
plot shows that inverted shapes occurred often just before a recession. Since the Great Recession
h
),
ce
7
4
i
Ed
r
e
of 2008, the Fed has kept short-term rates close to zero, which has resulted in unusually large
n
01
tio po
l
Fi 7. I
(
W
4
spreads to the long-term Treasury. Long-term rates have been coming down.
,
e
W
vo th
(
2
r
To understand these dynamics better, let’s work with a simpler two-year example. Let’s say
nc
a
o
7. The two possible
ce
that the yield curve tells you that the 1-year rate is r1 = 5% and the two-year rate is r2 = 10%.
)
(4
i
r
,C
explanations are (1) higher

F
a
01
o
t
You can work out that the 1-year forward rate is then r1,2 ≈ 15.24%. There are really only two
W
future interest rates and/or

u
n
e
7.
possible explanations:
h
(2) compensation for risk.

h
t
c
.
t
1
1. The 1-year interest rate next year will be higher than the 5% that it is today. Indeed, maybe
i
e
7
4
or
0
c
F
next year’s 1-year interest rate will be the 15.24% that it would be in a perfect world with
on
01
I
(
u
y2
perfect certainty. In equation form, is it the case that (1 + r1,2 ) · (1 + E(r2,3 )) = (1 + r1,3 ), so
h,

.
01 nce
J
h
i
7
elc iti
E(r2,3 ) = (1 + r1,3 )/(1 + r1,2 ) – 1? (I am using “E” as an abbreviation for “expected”—which

F
elc
7.
t
1
,
l
you will learn about in the next chapter.)

)
v
Ju
F
y 2 na
1
I
W
t
2. Investors tend to earn higher rates of return holding long-term bonds than they do holding
u
lch io
0
ra
7.
short-term bonds. For example, if the yield curve were to remain at exactly the same shape
h
i
or
F
t
an vo
o
next year, then a $100 investment in consecutive 1-year bonds would give you interest
di
l
or , Ju e Fi
u
p
,C
e
of only about $10.25, while the same investment in two-year bonds would give you (on
I
(
o W Ed Co , J
or
u
n
average) $21. Is E(r2,3 ) < (1 + r1,3 )/(1 + r1,2 ) – 1?

17 ce
i
or
at
n)
eF
In other words, the question is whether higher long-term interest rates today predict higher
4t
l
d
u
interest rates in the future, or whether they offer extra compensation for investors willing to
e
io
l
I
(
E
J
r
hold longer-term bonds. Let’s consider two possible variants of each of these two possibilities.
n
ra
t
.
,
di
h
i
r
)
y
Co , Ju e F
po
t
l
)
4
v
d
u
h,
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
o
n
01
ul
W
v
4
rp
h,
te
o
7.
20
a
I
(
,J
Recession
Recession
Recession
Recession
Recession
Recession
Recession
Recession
Recession
Recession
th
2
n
an vo
a
t
o
01
.
di
n
i
lch ion or
C
y
F
o
.I
na
ul
W
E
y2
Iv
15
p
,
e
o
Baa, Aaa, and GS20
a
ce
J
r
at
th
i
y2
in
o
c
F
t
o
.
),
i
l
Iv
r
C
d
e
20
F
e
o
l
W
10
n
E
t
Ju
p
,
n
a
7.
ch
a
or
th
ly
Fi
2
or
in
o
c
t
01

di
l
n
Iv
Ju
,C
ly
4
e
F
5
o
na
W
(
oW E
y2
Ju
p
io
,
ce
or
)
h
i
2
r
o
F
t
n
tio po
ul
4t
di
n
Iv
0
po ly
lch io
e
0
e
o
a
1960 1970 1980 1990 2000 2010

e(
E
2
or
at
Ju
p
in
o
.
Month
7
r
h
ly
i
or
C
c
c
t
o
n
01
Exhibit 5.4: 3-Month and 20-year Treasury Yields. The blue line is the yield on 3-month Treasury bills; the black line is
i
l
n
v
Ju
r
4
d
e
rp
,
I
na
the yield on 20-year Treasury bonds. When the blue line was above the black line, the yield curve was inverted. The gray
W
e(
2
o
a
line is the 12-month moving average inflation rate. Recessions are marked at the top (and compressed at the bottom).
,
i
r
)
ly
i
Ed
iti
r
The data comes from the Federal Reserve FRED database.
an vo
c
c
F
o
n
po
l
an
(4
d
lch io
e
20
te
n
h
W
h
or
t
n
elc
(4 elc itio
a
.
t
di
ce
(4
ly
r
C
c
n
7. e (4 elc itio po
W
l
u
Does the Yield Curve Predict Higher Future Inflation?
d
o
e
0
,
W
(
elc iti
2
t
n
vo
.
In general, when inflation is higher, you would expect investors to demand higher nominal
h
),
ce
7
If in ation is high, investors
4
i
Ed
r
e
interest rates. Consequently, you would expect nominal rates to go up when inflation rate
n
01
tio po
(typically) demand higher

Fi 7. I
(
W
expectations are going up. Similarly, you would expect nominal rates to go down when inflation
4
,
interest rates.
e
W
vo th
(
rate expectations are going down. Of course, demand and supply do not mean that real rates of
2
r
nc
a
o
7.
ce
return need to be positive—indeed, the real rate of return is often negative, but the alternative
)
(4
i
r
,C
F
a
01
of storing money under the mattress is even worse.
o
t
W
u
n
Therefore, our first potential explanation for an upward-sloping yield curve is that investors
7.
h
h
Are higher future in ation

t
c
believe that cash will be worth progressively less in the more distant future. That is, even though
a
.
t
1
rates the cause of higher

i
e
7
4
or
you will be able to earn higher interest rates over the long run, you may also believe that the
0
future interest rates?
c
F
on
01
I
(
inflation rate will increase from today’s rate. Because inflation erodes the value of higher interest
u
y2
p
h,

.
01 nce
rates, interest rates should then be higher in the future just to compensate you for the lesser value
h
i
7
elc iti
r
F
elc
7.
of money in the future. Of course, this argument would apply only to a yield curve computed
t
1
,
l
)
v
Ju
from Treasury debt that pays off in nominal terms. It should not apply to any bond payoffs that
F
y 2 na
1
I
W
t
are inflation-indexed.
lch io
0
ra
7.
h
h
i
or
Fortunately, since 1997 the Treasury has been selling bonds that are inflation-indexed. These
F
t
an vo
o
TIPS are in ation-indexed

di
l
bond contracts are written so that they pay out the promised interest rate plus the CPI inflation
4
or , Ju e Fi
u
p
,C
e
Treasury bonds. They are

I
(
rate. They are called Treasury Inflation Protected Securities (TIPS), or sometimes just CPI bonds.
o W Ed Co , J
or
not affected by in ation.

u
n
17 ce
By definition, they should not be affected by inflation in a perfect market. If the nominal yield
i
or
at
n)
eF
curve is upward-sloping because of higher future inflation rates, then a TIPS-based real yield
4t
l
d
u
curve should not be upward-sloping.

e
io
l
I
(
E
J
r
n
ra
t
.
,
di
h
i
r
)
y
Co , Ju e F
po
t
l
)
4
v
d
u
h,
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
5.4. Why Does the Yield Curve Usually Slope Up? 97
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
o
In ation-Neutral Bonds
01
ul
W
v
4
rp
h,
te
o
7.
a
I
(
,J
th
As it turns out, inflation-adjusted bonds had already been invented once before! The world’s first known inflation-indexed
2
n
an vo
a
t
o
01
.
bonds were issued by the Commonwealth of Massachusetts in 1780 during the Revolutionary War. These bonds were
di
n
i
lch ion or
C
y
F
o
.I
na
invented to deal with severe wartime inflation and discontent among soldiers in the U.S. Army with the decline in the
ul
W
E
y2
Iv
p
,
e
o
purchasing power of their pay. Although the bonds were successful, the concept of indexed bonds was abandoned after
a
ce
J
r
at
th
i
y2
in
the immediate extreme inflationary environment passed, and largely forgotten. In 1780, the bonds were viewed as an
o
c
F
t
o
.
),
i
l
Iv
r
irregular expedient, because there was no formulated economic theory to justify indexation.
C
d
e
20
F
e
o
l
W
n
E
Robert Shiller, “The Invention of Inflation-Indexed Bonds in Early America,” October 2003
t
Ju
p
,
n
a
7.
ch
a
or
th
ly
Fi
2
or
in
o
c
t
01

di
l
n
Iv
Ju
,C
ly
4
e
F
o
na
W
Conveniently, the Treasury website also shows a TIPS-based yield curve. They were (
oW E
y2
Ju
p
In ation-adjusted bond
io
,
ce
or
)
h
i
prices suggest expectations
2
r
o
F
t
Maturity 5-year 7-year 10-year 20-year 30-year
n
tio po
ul
of in ation were not the
4t
di
n
Iv
po ly
lch io
e
0
main driver of the

e
o
TIPS Interest Rate 0.45% 0.59% 0.73% 1.07% 1.28%
a
e(
E
2
or
at
upward-sloping yield curve.
Ju
p
in
o
7 .
r
h
ly
Recall that for small figures, the difference between the nominal and the real rate is about
i
or
C
c
c
t
o
n
01
ä Inflation Adjusting,
i
the inflation rate.
l
n
v
Ju
r
Formula 5.2, Pg.84.
d
e
rp
,
I
na
W
e(
2
o
Maturity 5-year 7-year 10-year 20-year 30-year
a

,
i
r
)
ly
i
Ed
iti
r
an vo
c
c
F
o
n
po
Ordinary Treasury Bonds 1.76% 2.09% 2.27% 2.67% 3.01%
l
an
(4
d
lch io
e
20
te
TIPS Interest Rate 0.45% 0.59% 0.73% I 1.07% 1.28%
n
h
W
h
or
t
n
elc
(4 elc itio
a
.
Implied Inflation 1.31% 1.50% 1.54% 1.60% 1.73%
t
di
ce
(4
ly
r
C
c
n
7. e (4 elc itio po
W
l
u
Without going into more details, the implied inflation rate contains a little bit of risk compen-
d
o
e
0
,
W
sation, too, and more for longer-term projects. Thus, differences in inflation expectations can
(
elc iti
2
t
n
vo
h
explain at most an 0.4% difference between 5-year and 30-year nominal interest rates, and likely
),
ce
7
4
i
Ed
r
e
quite a bit less. Say 0.25% is a good guestimate. This 0.25% is only about one-fifth part of the
n
01
tio po
l
Fi 7. I
(
W
4
1.25% spread in nominal interest rates. Trust me that these numbers have also been reasonably
,
e
W
vo th
(
2
r
representative for the last few decades of U.S. history. We can conclude that, even if increasing
nc
a
o
7.
ce
inflation expectations can play a minority role, they were not the main reason for the common
)
(4
i
r
,C
F
a
01
o
upward nominal yield curve slope.
t
W
u
n
e
7.
h
h
t
c
.
t
1
e
Q 5.20. In June, 2016, an inflation-adjusted 30-year Treasury bond offered a real yield of about
7
4
or
0
c
F
on
0.7% per year. The equivalent non-inflation-adjusted bond offered 2.25% per year. In what
01
I
(
u
y2
p
h,
inflation scenario would you be better off buying one or the other? (The most recent historical
.
01 nce
J
h
i
7
elc iti
inflation rate was 1% per year.)

F
elc
7.
t
1
,
l
)
v
Ju
F
y 2 na
1
I
W
t
u
lch io
0
ra
7.
Does The Yield Curve Predict Higher Future Interest Rates?

h
i
or
F
t
an vo
o
di
l
or , Ju e Fi
u
p
,C
A closely related possibility is that the yield curve is typically upward-sloping because short-term
e
Does a high forward

(
o W Ed Co , J
or
interest rates will be higher in the future. This is more generic than the previous explanation—
n
interest rate predict a high

17 ce
higher future interest rates need not be caused by higher future inflation expectations. Maybe
i
or
at
future interest rate?

n)
eF
4t
the 30-year yield of 3.01% was much higher than the 1-year yield of 0.65% because investors
l
d
u
e
io
expected the 1-year interest rate in 2044 to be much much higher than 3% (the forward rate,
n
l
I
(
E
J
r
r29,30 ). This does not tell you why investors would expect interest rates to be so much higher in
ra
t
.
,
di
h
i
r
)
2044 than in 2015—maybe they expect that capital will be more scarce then and investment
Co , Ju e F
po
t
l
opportunities will be better—but we can speculate about this even if we do not know the precise
)
4
v
d
u
h,
0
a
n
I
e(
reason.
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
Unfortunately, we do not have a direct estimate of future interest rates the way we had a
o
n
01
Alas, the historical data
ul
W
v
direct estimate of future inflation rates (from TIPS). Therefore, investigating this hypothesis
4
rp
h,
te
o
tells us “probably not much.”
7.
a
I
(
,J
requires looking at many years of evidence to learn whether future interest rates were well
th
2
n
an vo
a
t
o
01
.
predicted by prevailing forward rates. The details are beyond our scope. However, I can tell you
di
n
i
lch ion or
C
y
F
o
the punchline: Expectations of higher future rates of return are not the reason why the yield
.I
na
ul
W
E
y2
Iv
p
,
curve is typically upward-sloping (except maybe at the very short end of the yield curve, say, for
e
o
h
a
ce
J
r
at
th
i
y2
interest rates that are for cash investments for less than 1 month).
in
o
c
F
t
o
.
),
i
l
Iv
r
C
d
e
20
F
e
o
l
W
Does The Yield Curve Identify Bargains?
n
E
t
Ju
p
,
n
a
7.
ch
a
or
th
ly
Fi
2
or
in
o
If it is not the case that future interest rates are higher when forward rates are higher, it means
c
t
01
It must be either higher

di
l
n
Iv
Ju
that we are dealing with the second possible reason: On average, investors must have earned
,C
ly
4
e
F
future interest rates or
o
na
W
(
more in long-term bonds than in rolled-over short-term bonds. The empirical data confirms that
oW E
y2
higher compensation for
Ju
p
io
,
you would have ended up historically with more money if you had bought 30-year bonds than if
ce
long-term bond investors.
or
)
h
i
2
r
o
F
t
n
you had bought one-month bonds every month for 30 years.

tio po
ul
4t
di
n
Iv
po ly
lch io
e
0
e
One reason why this may have been the case is the habitat theory: Different investors
o
a
e(
E
Different Preferences?
2
or
at
Ju
p
in
may only like different types of bonds, to the point that different-maturity bond markets are
o
7 .
r
h
ly
i
or
segmented. The fact that there are habitats may well be true, but it is not so clear why long-term
C
c
c
t
o
n
01
i
l
bond investors would then be so scarce that they require more compensation, while short-term
n
v
Ju
r
4
d
e
rp
,
I
na
bond investors would be so abundant that bills can be sold for higher prices. And if this were
W
e(
2
o
a

the only reason, then why would borrowers not always prefer to sell short-term bonds instead?
,
i
r
)
ly
i
Ed
iti
r
an vo
c
c
And why would other investors not try to step in and get rich (by “arbitraging” the difference)?
F
o
n
po
l
an
(4
d
lch io
e
20
So why were long-term bonds better investments than short-term bonds? Maybe the yield
te
n
h
Free money? Not in a
W
h
or
curve was upward-sloping because investors were stupid. In this case, you might conclude
t
n
elc
(4 elc itio
perfect market.
a
.
t
di
ce
that the 30-year bond offering 3.01% was a much better deal than the 1-year bond offering
(4
ly
r
C
c
n
7. e (4 elc itio po
0.65%. Alas, investor stupidity seems highly unlikely as a good explanation. The market for
W
l
u
d
o
e
0
,
Treasury bond investments is close to perfect in the sense that we have used the definition.
J
W
(
elc iti
2
t
n
vo
It is very competitive. If there was a great deal to be had, thousands of traders would have
a
h
),
ce
7
4
i
Ed
immediately jumped on it. More likely, the interest rate differential does not overthrow the old
e
01
tio po
l
Fi 7. I
(
W
tried-and-true axiom, You get what you pay for. It is just a fact of life that investments for which
4
,
e
W
the payments are tied down to occur in 30 years must offer higher interest rates now in order
vo th
(
2
r
nc
a
o
7.
to entice investors—for some good reason yet to be identified. Again, it is important that you
ce
)
(4
i
r
,C
keep in mind that your cash and consumption are not tied down if you invest in a 30-year bond,
a
01
o
t
W
u
n
because you can, of course, sell your 30-year bond tomorrow to another investor if you so desire.
p
e
7.
h
h
t
c
.
t
1
e
7
Does It Compensate Investors For Risk?
4
or
0
c
F
on
01
I
(
u
y2
p
h,
If it isn’t market stupidity that allows you to earn more money in long-term bonds than in
.
01 nce
The answer is probably

h
i
7
elc iti
rolled-over short-term bonds, then what else could it be? The empirical evidence suggests that it
F
elc
7.
compensation for risk.
t
1
,
l
is most likely the phenomenon explained in Section 5.3: Interim changes in prevailing interest
v
Ju
F
y 2 na
1
I
rates have much more impact on long-term bonds than they have on short-term bonds. Recall
W
t
u
lch io
0
ra
ä Bond Risk,
7.
that rolling over short-term bonds insulates you from the risk that interest rates will change
h
h
i
Sect. 5.3, Pg.93.

or
F
t
an vo
o
in the future. If you hold a one-day bond and interest rates double by tomorrow, you can just
t
di
l
or , Ju e Fi
u
p
,C
e
purchase more bonds tomorrow that will offer you twice the interest rate. In contrast, if you hold
I
(
o W Ed Co , J
or
a long-term bond, you could lose your shirt if interest rates go up in the future. With long-term
n
17 ce
i
bonds being riskier than short-term bonds, investors only seem to want to buy them if they get
or
at
n)
eF
4t
some extra rate of return. Otherwise, they prefer rolling short bonds. Thus, long-term bonds
l
d
u
e
io
need to offer investors more return on average than short-term bonds.

n
l
I
(
E
J
r
n
ra
t
.
,
di
h
i
r
)
y
Co , Ju e F
po
t
l
)
4
v
d
u
h,
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
5.5. Corporate Time-Varying Costs of Capital 99
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
5.5 Corporate Time-Varying Costs of Capital
y
d
o
n
01
ul
W
v
4
rp
h,
te
o
7.
a
I
Now that you understand that the yield curve is usually upward-sloping for a good reason, you
(
,J
th
2
Extend this insight to
n
an vo
a
t
o
01
should recognize the family resemblance: Corporate projects are offering cash flows, just like
.
di
n
corporations: Longer-term
i
lch ion or
C
y
F
Treasury bonds. Thus, it should not surprise you that longer-term projects usually have to offer
o
.I
projects, even if they are

na
ul
W
E
y2
Iv
higher rates of return than shorter-term projects. And just because a longer-term project offers a
p
,
not more likely to default,
e
o
h
a
ce
J
r
at
th
often face a higher cost of
i
higher expected rate of return does not necessarily mean that it has a higher NPV. Conversely,
y2
in
o
c
F
t
o
.
capital, and therefore
),
just because shorter-term borrowing allows firms to pay a lower expected rate of return does not
i
l
Iv
r
C
d
e
20
F
should have to deliver
e
o
necessarily mean that this creates value. (Neither firms nor the U.S. Treasury rely exclusively on
l
W
n
E
t
higher returns.
Ju
p
,
n
a
short-term borrowing.) Paying a higher expected rate of return for longer-term obligations is
7.
ch
a
or
th
ly
Fi
2
or
in
(usually) a fact of life.
o
c
t
01

di
l
n
Iv
Ju
,C
ly
4
e
F
o
na
IMPORTANT
W
(
oW E
y2
Ju
p
io
Even in a perfect market without uncertainty:
,
ce
or
)
h
i
2
r
o
F
t
n
• The appropriate cost of capital (rate of return) should usually depend on how long-term
tio po
ul
4t
di
n
Iv
po ly
lch io
e
0
the project is.

e
o
a
e(
E
2
or
at
Ju
p
in
o
• The term structure is usually upward-sloping. Short-term corporate projects usually have
7 .
r
h
ly
i
or
C
lower costs of capital than long-term projects.
c
c
t
o
n
01
i
l
n
v
Ju
r
4
d
e
rp
,
• Conversely, corporations usually face lower costs of capital (expected rates of return offered
I
na
W
e(
2
to creditors) if they borrow short term rather than long term.
o
a

,
i
r
)
ly
i
Ed
iti
r
The difference between long-term and short-term rates is called the Term Premium.
an vo
c
c
F
o
n
po
l
an
(4
d
lch io
e
20
te
n
h
W
h
or
t
n
elc
(4 elc itio
a
.
t
Let me give you a short preview now. In Chapter 10, you will learn about the CAPM. The
di
ce
(4
ly
Time-Varying Expected
r
C
c
CAPM is the most common model used to discount future cash flows in NPV applications. It is
n
7. e (4 elc itio po
Rates of Return vs. Time-

W
l
u
d
a model that relates your project’s required expected rate of return to its risk. In practice, the
o
e
0
,
and Risk-Varying Expected

J
W
(
elc iti
2
t
CAPM allows you to use higher (risk-free) rates of return for cash flows farther in the future. Rates of Return.
n
vo
h
),
ce
7
Thus, the CAPM has one term that is (more or less reflective of) the term-premium and one
4
i
Ed
r
e
01
tio po
l
Fi 7. I
term for the risk-premium. In this sense, it can be viewed as a generalization of the point of this
(
W
4
,
chapter that longer-term projects usually require higher (opportunity) costs of capital. If the
e
W
vo th
(
2
r
nc
second term is zero, then all you have left is the term premium.
a
o
7.
ce
)
(4
i
r
,C
It turns out that the first term (the term premium) has worked much better than the second
F
a
01
o
Basic Usage and Reasonable
t
W
(the risk premium). Thus, in real life, it is often more important for you to understand that you
n
guesstimate.
e
7.
h
h
usually need to increase your cost-of-capital estimate for longer-term cash flows, than it is for
t
c
.
t
1
you to understand the much more complex second CAPM term.

i
e
7
4
or
0
c
F
on
01
Alas, there is one second-order complication: the term-premium for corporate cash flows
I
(
u
y2
p
h,
can be different from the term-premium for Treasuries. Although this is true, the first-order
.
01 nce
J
h
i
7
elc iti
r
F
approximation is usually that the two are similar. That is, if a 20-year Treasury offers an expected
elc
7.
t
1
,
l
)
v
rate of return that is 2% higher than that of a 1-year Treasury, you should probably guess that a
Ju
F
y 2 na
1
I
W
t
corporate cash flow in 20 years should offer an expected rate of return that is also about 2%
u
lch io
0
ra
7.
higher than that of its 1-year counterpart. (The second-order corrections depend on much deeper,
h
i
or
F
t
an vo
o
more difficult, and harder-to-prove reasoning and will most likely gross up or shade down the
t
di
l
or , Ju e Fi
u
p
,C
e
Treasury spread only by a little bit.)

I
(
o W Ed Co , J
or
u
n
17 ce
i
or
at
n)
eF
4t
l
d
u
e
io
l
I
(
E
J
r
n
ra
t
.
,
di
h
i
r
)
y
Co , Ju e F
po
t
l
)
4
v
d
u
h,
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
Summary
y
d
o
n
01
ul
W
v
4
rp
h,
te
o
7.
a
I
(
,J
This chapter covered the following major points: • The relationship between nominal interest rates, real
th
2
n
an vo
a
t
o
01
.
interest rates, and inflation rates is

di
n
• Different horizon investments can offer different rates
i
lch ion or
C
y
F
o
.I
na
ul
W
of return. This phenomenon is often called time-
E
y2
Iv
p
,
e
o
varying rates of return. (1 + rnominal ) = (1 + rreal ) · (1 + Inflation Rate)
a
ce
J
r
at
th
i
y2
in
o
c
F
t
o
.
• The general formula for compounding works just as
),
i
l
Iv
r
C
d
e
20
F
e
well for time-varying rates of return as it does for
o
l
W
• Unlike nominal interest rates, real interest rates can—
n
E
t
Ju
time-constant rates of return. You only lose the abil-
p
,
n
a
7.
ch
a
and often have been—negative.
or
th
ly
Fi
ity to exponentiate (one plus the 1-year rate of return)
2
or
in
o
c
t
01

di
when you want to compute multiyear rates of return. • In NPV, you can either discount real cash flows with
l
n
Iv
Ju
,C
ly
4
e
F
o
na
real interest rates, or discount nominal cash flows
W
• A holding rate of return can be annualized for easier (
oW E
y2
Ju
p
io
with nominal interest rates. The latter is usually more
,
interpretation.
ce
or
)
h
i
2
r
convenient.
o
F
t
n
tio po
ul
• The graph of annualized interest rates as a function
4t
di
n
Iv
po ly
lch io
e
0
e
of maturity is called the “term structure of interest • TIPS are Treasury bonds that protect against future
o
a
e(
E
2
or
at
Ju
p
in
rates” or the “yield curve.” inflation. Short-term bond buyers are also less ex-
o
7 .
r
h
ly
i
or
posed to inflation rate changes than long-term bond

C
• The yield curve is usually upward-sloping. However,
c
c
t
o
n
01
i
l
buyers.
n
v
Ju
r
no law of finance is violated if it is downward-sloping
d
e
rp
,
I
na
W
e(
(inverted), humped, or flat.
2
• Higher long-term interest rates could be due either
o
a

,
i
• Net present value also works just as well for time- to expectations of higher future interest rates or to
r
)
ly
i
Ed
iti
r
an vo
c
c
F
o
n
po
varying interest rates. You merely need to use the extra required compensation for investors willing to
l
an
(4
d
lch io
e
20
te
appropriate rate of return as the opportunity cost of Ihold longer-term bonds. The empirical evidence sug-
n
h
W
h
or
t
capital in the denominator. gests that historically the latter has been the more
elc
(4 elc itio
a
.
t
di
ce
(4
ly
r
C
important factor.
c
• An important side observation: “Paper losses” are no
n
7. e (4 elc itio po
W
l
u
4
different from real losses.
d
o
e
0
,
• Corporations should realize that corporate project

J
W
(
elc iti
2
t
• Inflation is the process by which money will buy fewer cash flows need to be discounted with specific costs
n
vo
h
),
ce
7
4
goods in the future than it does today. If contracts of capital that may depend on when the cash flows
y
i
Ed
r
e
01
tio po
l
Fi 7. I
are inflation-indexed in a perfect market, inflation is will come due. It is not unusual that cash flows in the
W
4
,
e
irrelevant. This is rarely the case. more distant future require higher discount rates.
h
W
vo th
(
2
r
nc
a
o
7.
ce
)
(4
i
r
,C
F
a
01
o
t
W
u
n
e
7.
Preview of the Chapter Appendix in the Companion

h
t
c
.
t
1
e
7
4
or
0
c
F
The appendix to this chapter explains

on
01
I
(
u
y2
p
h,

.
01 nce
• how you can translate the yield curve (in Exhibit 5.3) step by step into forward and total
J
h
i
7
elc iti
r
F
elc
holding rates of return.
7.
t
1
,
l
)
v
Ju
F
• how shorting works in the real world, and how you can lock in a future interest today with
y 2 na
1
I
W
t
u
lch io
0
clever bond transactions today.
ra
7.
h
h
i
or
F
• (again) how the “duration” of bonds helps you measure when you receive your cash flows
an vo
o
di
l
or , Ju e Fi
u
p
on average.
,C
e
I
(
o W Ed Co , J
or
• continuous compounding, which is a different way of quoting interest rates.

n
17 ce
i
or
at
n)
• that Treasury notes and bonds are not really zero bonds (as we pretended in this chapter)
eF
4t
l
but coupon bonds, and why this rarely matters in a corporate context. True Treasury zero
v
d
u
e
io
l
I
(
bonds are called STRIPS.

E
J
r
n
ra
t
.
,
di
h
i
r
)
y
Co , Ju e F
po
t
l
)
4
v
d
u
h,
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
End of Chapter 5 Material 101
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
o
n
01
Keywords
ul
W
v
4
rp
h,
te
o
7.
a
I
(
,J
th
2
n
an vo
a
t
o
Annualized rate, 77. Average rate of return, 77. BLS, 82. Bureau of Labor Statistics, 82. CPI bond, 96. CPI, 82.
01
.
di
n
i
lch ion or
C
y
Consumer Price Index, 82. Deflation, 82. Duration, 80. Forward rate, 76. GDP Deflator, 82. Great Recession, 83.
o
.I
na
ul
W
E
y2
Hyperinflation, 83. Inflation, 82. Inflation-indexed terms, 82. Long bond, 87. Macauley Duration, 80. Nominal
Iv
p
,
e
o
h
a
ce
J
return, 83. Nominal terms, 82. PPI, 82. Paper loss, 95. Producer Price Index, 82. Real return, 83. Real
at
th
i
y2
in
o
c
F
t
o
.
terms, 82. Reinvestment rate, 76. Spot rate, 76. T-bill, 86. TIPS, 96. Term Premium, 99. Term structure of
),
i
l
Iv
r
C
d
e
20
F
e
interest rates, 87. Term structure, 87. Treasuries, 87. Treasury bill, 86. Treasury bond, 86. Treasury note, 86.
o
l
W
n
E
t
Ju
p
,
n
Yield curve, 87.

a
7.
ch
a
or
th
ly
Fi
2
or
in
o
c
t
01

di
l
n
Iv
Ju
,C
ly
4
e
F
o
na
W
(
oW E
y2
Answers
Ju
p
io
,
ce
or
)
h
i
2
r
o
F
t
n
tio po
ul
4t
di
n
Iv
po ly
Q 5.1 r0,2 = (1 + r0,1 ) · (1 + r1,2 ) – 1 = 1.02 · 1.03 – 1 = 5.06%
lch io
e
0
e
o
a
e(
E
2
or
at
Ju
$100 $300 $500
p
Q 5.2 Solve (1 + x) · (1 + 22%) = (1 – 50%), so the project had a
in
o
.
– $200 + + + ≈ $612.60
7
r
1.042 1.0453
h
ly
rate of return of –59.00%. 1.03
i
or
C
c
c
t
o
n
01
i
l
n
v
Ju
r
4
Q 5.3 The first three-year compounded rate of return was
d
e
rp
,
I
na
W
e(
r2010,2012 ≈ (1 + 0.150) · (1 + 0.021) · (1 + 0.16) – 1 ≈ +36.2%.
2
Q 5.12 The CPI is the average price change to the consumer for
o
a

,
(The notation is a bit ambiguous when month and day are omit-
i
a specific basket of goods. The PPI measures the price that pro-
r
)
ly
i
Ed
iti
r
an vo
c
c
F
ted, because the first return is from the end of 2009 to the end
o
ducers are paying. Taxes, distribution costs, government subsidies,
n
po
l
an
(4
of 2010.) The second three-year rate was r2013,2015 ≈ +52.42%.
d
and basket composition drive a wedge between these two inflation
lch io
e
20
te
n
h
W
h
The full six-year compounded rate of return was thus r2010,2015 ≈ measures.
or
t
n
elc
(4 elc itio
a
.
t
(1+.362%)·(1+.524%)–1 ≈ +107.6%. Although these were very fat

di
ce
Q 5.13 (1 + rnominal ) = (1 + rreal ) · (1 + π)
(4
ly
r
C
c
years for stock investors. the realized rate was indeed time-varying.
n
7. e (4 elc itio po
W
l
u
Q 5.14 1.20/1.05 ≈ 1.1429. The real interest rate is 14.29%.
d
o
e
0
,
Q 5.4 The returns were (–33%, +50%, –67%, +100%). Thus the
J
W
(
elc iti
Q 5.15 The nominal interest rate is 1.03 · 1.08 – 1 = 11.24%.

t
average rate was 12.5% and the overall rate of return was –33.33%.
n
vo
h
),
ce
Therefore, the cash flow is worth about $500,000/1.1124 ≈
7
4
It is always true that the compound rate of return is always less than
Ed
r
e
$449,479.
n
01
tio po
l
Fi 7. I
the average rate of return. The example shows that the two can
W
4
,
differ in sign. Q 5.16 Bills, notes, and bonds. T-bills have maturities of less
e
W
vo th
(
2
r
nc
than 1 year. T-notes have maturities from 1 to 10 years. T-bonds

a
o
Q 5.5 1.05 12/4

≈ 15.76%
7.
ce
)
(4
have maturities greater than 10 years.

i
r
,C
p
F
a
01
o
t
W
Q 5.6 The annualized rate of return is 1.4 – 1 ≈ 18.32%. It is

u
Q 5.17 Yes. The answers are right in the table. The three-
n
e
7.
therefore lower than the 20% average rate of return. year rate of return is 1.01313 – 1 ≈ 3.98%. The forward rate is
h
h
t
c
.
1.0398/(1.0065 · 1.0147) – 1 ≈ 1.81%.
t
1
e
Q 5.7 The compounded rate of return is always higher than the
7
4
or
0
c
F
sum, because you earn interest on interest. The annualized rate Q 5.18 r0,5 = 1.01765 – 1 ≈ 9.12%. Therefore, 1 + r3,5 =
on
01
I
(
u
y2
of return is lower than the average rate of return, again because 1.01765 /1.01313 – 1 ≈ 4.94%. This is a two-year forward hold-
p
h,

.
01 nce
J
h
i
7
elc iti
you earn interest on the interest. For example, an investment of ing rate of return. Thus, it is 1.0494( 1/2) – 1 ≈ 2.44% in annualized
F
elc
7.
t
1
,
l
$100 that turns into an investment of $200 in two years has a terms.
)
v
Ju
F
y 2 na
total holding rate of return of 100%—which is an average rate

n
1
I
W
t
Q 5.19 (1) For the 1-year bond, the value of a $100 bond changes
u
of return of 100%/2 = 50% and an annualized rate of return of

lch io
0
ra
p
7.
from $100/1.0800 ≈ $92.59259 to $100/1.0801 ≈ $92.58402.

h
h
i
(1 + 100%) – 1 ≈ 41.42%. Investing $100 at 41% per annum

or
F
t
an vo
This is about a –0.009% change. (2) For the ten-year bond, the
o
di
l
would yield $200, which is lower than 50% per annum.

value of a $100 bond changes from $100/1.0810 ≈ $46.31935 to
4
or , Ju e Fi
u
p
,C
e
I
(
$100/1.080110 ≈ $46.27648. This is a –0.09% change—ten times

o W Ed Co , J
or
Q 5.8 The six-year holding rate of return was 107.6%. Thus, the
u
n
p
17 ce
6
annualized rate of return was r8 = 1 + 107.6% – 1 ≈ 12.9%. that of the 1-year bond. (3) The derivative of the 1-year bond is
i
or
at
n)
eF
–0.009/0.01 = –0.9 ≈ –1. The derivative of the ten-year bond is

4t
l
1/5
Q 5.9 r0,5 = 50% (1 + r5 )5 = 1.50
v
d
u
=⇒ r5 = 1.50 –1 ≈ –0.09/0.01 ≈ –9. The derivative of the ten-year bond is about nine
e
io
l
I
(
E
J
r
8.45%. times more negative.

t
n
ra
t
.
,
di
Q 5.20 If the inflation rate will increase to more than

i
Q 5.10 The annualized five-year rate of return is the same 10%.

)
y
Co , Ju e F
po
1.0225/1.007 – 1 ≈ 1.5% per year, the inflation-adjusted bond will

l
)
4
v
d
u
h,
Q 5.11 This project is worth

0
be better. Otherwise, the non-inflation adjusted bond will be better.

a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
End of Chapter Problems
y
d
o
n
01
ul
W
v
4
rp
h,
te
o
7.
a
I
(
,J
th
2
Q 5.21. Are you better off if a project first returns –10% followed Q 5.29. A project has cash flows of +$100 (now at time 0), and
n
an vo
a
t
o
01
.
di
by +30%, or if it first returns +30% followed by –10%? –$100, +$100, and –$100 at the end of consecutive years. The
n
i
lch ion or
C
y
F
o
.I
interest rate is 6% per annum.

na
ul
W
E
y2
Iv
Q 5.22. Compare two stocks. Both have earned 8% per year on
p
,
e
o
h
a
ce
J
average. However, stock A has oscillated between 6% and 10%.
at
th
i
y2
1. What is the project’s NPV?
in
o
c
F
t
o
.
Stock B has oscillated between 3% and 13%. (For simplicity, say
),
i
l
Iv
r
C
d
2. How does the value change if all cash flows will occur one
e
20
F
e
that they alternated.) If you had bought $500 in each stock, how
o
l
W
n
E
year later?
t
Ju
much would you have had 10 years later?
p
,
n
a
7.
ch
a
or
th
ly
Fi
3. Repeat these two questions, but assume that the 1-year
2
or
in
o
c
t
Q 5.23. (Strange) Stock A always alternated between +20% and
01

(annualized) interest rate is 5%, the two-year is 6%, the
di
l
n
Iv
Ju
,C
ly
4
–10% in the past. Stock B earned 4.5% per annum.
F
three-year is 7%, the four-year is 8%, and so on.
o
na
W
(
oW E
y2
Ju
p
io
,
ce
1. What was the average rate of return for stock A?
or
)
h
i
2
r
Q 5.30. What is the current inflation rate?

o
F
t
n
tio po
ul
4t
2. What was the average rate of return for stock B?
di
n
Iv
po ly
lch io
e
0
e
Q 5.31. What is the annualized current nominal interest rate on
o
a
3. On a 1-year basis, would a risk-neutral investor prefer e(
E
2
or
at
Ju
30-day U.S. Treasury bills?
p
in
o
.
+20% or –10% with equal probability, or 4.5% for sure?

7
r
h
ly
i
or
C
c
c
t
o
Q 5.32. Using the information from Questions 5.30 and 5.31,
n
01
4. How much would each dollar invested 10 years ago in
i
l
n
v
Ju
r
4
d
compute the annualized current real interest rate on 30-day Trea-
e
rp
,
stock A have earned?

I
na
W
e(
2
suries.
o
5. How much would each dollar invested 10 years ago in
a

,
i
r
)
ly
i
Ed
iti
r
stock B have earned? Q 5.33. If the nominal interest rate is 7% per year and the inflation
an vo
c
c
F
o
n
po
l
an
rate is 2% per year, what is the exact real rate of return?
(4
d
6. What is going on here?
lch io
e
20
te
n
h
W
h
or
t
Q 5.34. The inflation rate is 1.5% per year. The real rate of return
elc
(4 elc itio
a
.
t
di
Q 5.24. Return to Question 5.3. What was the annualized geo-

ce
(4
ly
is 2.0% per year. A perpetuity project that paid $100 this year
r
C
c
n
metric rate of return, and what was the average rate of return on
7. e (4 elc itio po
will provide income that grows by the inflation rate. Show what
W
l
u
d
o
e
0
,
the S&P 500? Would stock brokers prefer to tell their clients the this project is truly worth. Do this in both nominal and real terms.
J
W
(
elc iti
2
t
former or the latter?

n
(Be clear on what never to do.)

vo
h
),
ce
7
4
i
Ed
r
e
Q 5.25. On June 23, 2016, the Brits voted to exit the EU. The
n
01
tio po
Q 5.35. If the annualized rate of return on insured tax-exempt

l
Fi 7. I
(
W
following were the daily values of an investment (in a fund called
4
,
municipal bonds will be 3% per annum and the inflation rate

e
W
SPY):
vo th
(
2
r
remains at 2% per annum, then what will be their real rate of

nc
a
o
7.
ce
return over 30 years?
)
(4
i
r
,C
F
a
01
o
t
June 27 28
W
u
n
Q 5.36. If the real interest rate is –1% per annum and the infla-
p
e
7.
Dollars 199.60 203.20

h
tion rate is 3% per annum, then what is the present value of a

t
c
.
t
1
$1,000,000 nominal payment next year?

i
e
7
4
or
0
c
F
on
01
I
(
If returns were to accumulate at the same rate over an entire year

E
Q 5.37. Inflation is 2% per year; the interest rate is 8% per year.

u
y2
p
h,
(252 trading days), what would a $100 investment turn into?

.
01 nce
Your perpetuity project has cash flows that grow at 1% faster than
J
h
i
7
elc iti
r
F
elc
inflation forever, starting with $20 next year.
7.
t
1
,
l
Q 5.26. If the annualized five-year rate of return is 10%, what is

)
v
Ju
F
y 2 na
the total five-year holding rate of return?

1
I
W
t
1. What is the real interest rate, both accurate (the “1+” ver-
lch io
0
ra
7.
Q 5.27. If the annualized five-year rate of return is 10%, and if sion) and approximate (the subtraction version)?
h
i
or
F
t
an vo
o
the first year’s rate of return is 15%, and if the returns in all other
t
2. What is the correct project PV?

di
l
or , Ju e Fi
u
p
,C
e
years are equal, what are they? 3. What would you get if you grew a perpetuity project of $20
I
(
o W Ed Co , J
or
u
n
by the real growth rate of 1%, and then discounted it at

17 ce
Q 5.28. The annual interest rate from year t to year t + 1 is

i
or
at
the nominal cost of capital?

n)
eF
rt,t+1 = 5% + 0.3% · t (e.g., the rate of return from year 5 to year

4t
l
4. What would you get if you grew a perpetuity project of $20

v
d
u
6 is 5% + 0.3% · 5 = 6.5%).
e
io
l
I
by the nominal growth rate of 3%, and then discounted it

(
E
J
r
n
ra
t
at the real cost of capital?

.
1. What is the holding rate of return of a ten-year investment

,
di
h
i
r
)
y
Co , Ju e F
po
today?
t
l
)
4
v
Performing either of the latter two calculations is not an uncom-

d
u
h,
2. What is the annualized interest rate of this investment?

a
n
I
e(
mon mistake in practice.

J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
t
17 ce ( We diti
y2
Co
in
h,
tio
ra
o
.
ce
,
17
th
,C
n)
eF
vo
elc
ul
di
an
E
h,
(4
C
tio
20
lch
.I
J
W
oW E
at
th
an elc
h,
o
ce
,
i
h
Fi
r
C
n)
ly
(4
c
o
t
End of Chapter 5 Material 103
n
E
l
v
4
u
p
h,
e
te
0
a
I
ce
(
J
E
or
h
2
n
o
(4 elc
ra
7.
4t
,
Iv
th
Fi
C
y
d
Q 5.38. You must value a perpetual lease. It will cost $100,000 Q 5.44. At today’s prevailing Treasury rates, how much money
o
n
01
ul
W
v
each year in real terms—that is, its proceeds will not grow in would you receive from an investment of $100 in 1 year, 10 years,
4
rp
h,
te
o
7.
a
I
(
real terms, but just contractually keep pace with inflation. The and 30 years? What are their annualized rates of return? What
,J
th
2
n
an vo
a
t
o
01
prevailing interest rate is 8% per year, and the inflation rate is are their total holding rates of return?
.
di
n
i
lch ion or
C
y
2% per year forever. The first cash flow of your project next year
o
.I
na
ul
W
E
y2
is $100,000 quoted in today’s real dollars. What is the PV of the
Iv
p
,
Q 5.45. Do long-term bonds pay more than short-term bonds
e
o
h
a
ce
project? (Warning: Watch the timing and amount of your first
J
r
at
th
because you only get money after a long time—money that you
i
y2
in
o
c
F
t
o
payment.)
.
could need earlier?
),
i
l
Iv
r
C
d
e
20
F
e
o
Q 5.39. If the real rate of return has been about 1% per month
l
W
n
E
t
Ju
p
,
n
for long-term bonds, what would be the value of an investment Q 5.46. A five-year, zero-coupon bond offers an interest rate of
a
7.
ch
a
or
th
ly
Fi
2
that costs $100 today and returned $200 in 10 years?
or
8% per annum.
in
o
c
t
01

di
l
n
Iv
Ju
,C
Q 5.40. At your own personal bank, what is the prevailing savings
ly
4
e
F
o
na
W
account interest rate? (
oW E
y2
Ju
1. How does a 1-basis-point increase in the prevailing interest
p
io
,
ce
Q 5.41. Look up today’s yield curve on a financial website. What rate change the value of this bond in relative terms?
or
)
h
i
2
r
o
F
t
n
tio po
ul
is the 1-year rate of return on a risk-free Treasury? What is the
4t
di
n
2. What is the ratio of the relative bond value change over

Iv
po ly
lch io
e
0
ten-year rate of return on a risk-free Treasury? What is the 30-year

e
o
a
the interest change? (This is the derivative of the value

e(
E
2
or
at
rate of return on a risk-free Treasury?
Ju
p
in
o
with respect to interest rate changes.)
7 .
r
h
ly
i
or
C
Q 5.42. The 1-year forward interest rates are
c
c
t
o
n
3. How does the derivative of wealth with respect to the in-

01
i
l
n
v
Ju
r
4
d
e
terest rate vary with the length of the bond?
rp
,
I
na
Y1 Y2 Y3 Y4 Y5 Y6
W
e(
2
o
a

,
i
3% 4% 5% 6% 6% 6%
r
)
ly
i
Ed
iti
r
Q 5.47. Look at this week’s interest rate on ordinary T-bonds and
an vo
c
c
F
o
n
po
l
an
Y7 Y8 Y9 Y10 Y11 Y12 on TIPS. (You should be able to find this information, e.g., in the
(4
d
lch io
e
20
te
n
h
Wall Street Journal or through a fund on the Vanguard website.)
W
h
or
7% 7% 7% 6% 5% 4%
t
n
elc
(4 elc itio
a
.
What is the implied inflation rate at various time horizons?
t
di
ce
(4
ly
r
C
c
n
7. e (4 elc itio po
1. Compute the 12 n-year compounded holding rates of return

W
l
u
4
Q 5.48. The yield curve is usually upward-sloping. Assess whether
d
o
e
0
,
from now to year n.

J
W
(
this means that the following statements are true or false:
elc iti
2
t
n
vo
2. Compute the 12 annualized rates of return.

a
h
),
ce
7
4
i
Ed
r
e
3. Draw the yield curve.

n
01
tio po
l
Fi 7. I
(
W
1. Investors earn a higher annualized rate of return from
4
4. Is there anything wrong in this example?
,
e
long-term T-bonds than short-term T-bills.

h
W
vo th
(
2
r
nc
Q 5.43. The annualized interest rates are

a
o
7. 2. Long-term T-bonds are better investments than short-term
ce
)
(4
i
r
,C
T-bills.
a
01
o
t
W
Y1 Y2 Y3 Y4 Y5 Y6
u
n
3. Investors are expecting higher inflation in the future than

7.
h
h
3% 4% 5% 6% 6% 6%
c
they are today.

a
.
t
1
e
7
4
or
0
Y7 Y8 Y9 Y10 Y11 Y12 4. Investors who are willing to take the risk of investing in
c
F
on
01
I
(
u
y2
long-term bonds on average earn a higher rate of return

p
h,
7% 7% 7% 6% 5% 4%
.
01 nce
because they are taking more risk (that in the interim bond
h
i
7
elc iti
r
F
elc
7.
t
prices fall / interest rates rise).

1
,
l
)
v
1. Draw the yield curve.

Ju
F
y 2 na
1
I
W
t
2. Compute the 12 n-year compounded holding rates of return

u
lch io
0
ra
7.
Q 5.49. Evaluate and Discuss: Does the evidence suggest that

h
from now to year n.

h
i
or
F
long-term bonds tend to earn higher average rates of return than

an vo
o
3. Compute the 12 1-year forward rates of return.

di
l
or , Ju e Fi
short-term bonds? If yes, why is this the case? If no, why is this
u
p
,C
e
4. Is there anything wrong in this example?

I
(
not possible?
o W Ed Co , J
or
u
n
17 ce
i
or
at
n)
eF
4t
l
d
u
e
io
l
I
(
E
J
r
n
ra
t
.
,
di
h
i
r
)
y
Co , Ju e F
po
t
l
)
4
v
d
u
h,
0
a
n
I
e(
J
or
W
y2
n
c
p
7.
,
l
i
i
C
n)
e
vo
W
d
rp
,
,C
te
elc tio
W
th
in
n) ora
Ed Co
.
ce
di
17
y
(4
Ju e F
i
l
n
E
Ed
u
p
h,
na
ce
,J
r
h
iti
7.
ce
Iv
th
Fi
or
ly
01
W
e(
Iv
rp
te
e
io
7.
h
ce W di or Ju
l Fi 7. I (4 lch io
n or y2 an elc
7. (4 e l c t io p o n a vo t h , ) a t 01 ce h,
Iv th h, n) ra
y2
0 nc W Ed C , Ju e F 7. (4 Co
t e 1 e i
or
ly i Iv th
o W Ed Co , J
i r u l F i
7. ( 4
elc
h
tio po
r
na
n o E
e( n 20
4t elch tion pora y 20 nan Ivo th E , Co ), J ate 17 ce ( We diti
h , C ) , t e 1 c e W di Fi lch
W E di or J u F 7 . (4 e t rp
or
ul
y2 n an vo
.I 4t
h ,C
elc tio po l i I vo th
lch io
, n ) at 01 ce W E d o
h h, n) ra
y 2 na
W Ed C , Ju e F 7. (4 elc i tio
, J t 01 nce or
ly i Iv th h, n)
Ed Co u l
eF
i
7. ( 4
elc iti
h o po
r 2
na
n o E C ,
iti rp y n I v t h , n a 0 c W d o
ly in Iv th h, n) ra
Co , Ju e F
r l i
7. e (4 elc itio po
I h n r 2 a n o E C , t
io p y2 n t h ,C ), a te 01 ce W di or Jul
n) ora 0
an vo
W E d o J u F 7 . ( 4 e l c t io p o
,J te 17 ce i r ly i Iv th h n r
y2
u F i . ( 4
elc tio po
r
na
n o E , C ) , at
rp l y n a I v t h h, n ) a t
20 c W d o J u
eF
or e( el i t i rp
o ly in
p l n I 4t ch o n ra 20
Ju e F 7. ( elc iti o r y 2 an v h , C ) te
ly in I v 4t h on a 0 c oW E o ,J 17
.
20 anc o W h E , Co ), J te F 17 e( e di
t r p ul
y
Fi
n
te 17 e di
t r p u l i n . I 4 t
lch ion or
a 2 0
an
.
e(
4 or y 2 an vo h ,C ), te 1 ce
Fi
na I vo t h
lch io
,C n ) a t 01 ce W E di or J Fi 7
n W E , J e F 7 e l t p o
ul
y n .I
01 ce d or u i . (4 c io r 2 a n
vo
7. at 01 ce
01 ce W d or Ju e F 7. (4
W F ( e l i t i p o l in Iv th
ce d or u 7. 4 c o y2 a o
(4 elch itio por ly 2 inan Ivo th E h, C n), rate 017 nce W
Iv n 01 ce W d o Ju F . ( 4
elc
o W th E , Co ), J ate 7 ( e iti r p ly in I v t h h,
d r u F . 4 l c o o 2 a oW E
(4 elch itio por ly 2 inan Ivo th E h, C n), rate 017 nce e di
th n ) at 0 c e W d o Ju F . ( 4 l c tio
,C 1 r po ly in Iv th h,
e
Ed
i
or , Ju e Fi
p l
7. (4 elc itio
t h n r 2 a n o E C

Yield Curve

Caricato da

Informazioni sul documento

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Yield Curve

Caricato da

Copyright:

Formati disponibili

t

(4 elch itio por ly 2 nan Ivo th E h, C n),

at 201 nce o W Ed Cor , Ju e Fi 17.

vary with the length of time an investment requires. I

In Percent Per Annum

1m 3m 6m 1y 2y 3y 5y 7y 10y 20y 30y

an Ivo th E , C n), J ate

does inflation matter?

periods, surely so will firms!

U.S. Treasury yield curve.

based on the length of the

(4 elch itio por ly 2 nan Ivo th E h, C n),

using the same compounding formula as before:

(1 + r0,2 ) = (1 + 20%) · (1 + 30%) = (1 + 56%)

(1 + r0,1 ) · (1 + r1,2 ) = (1 + r0,2 )

compounding over many

Given: r1,2 = 30% r2,3 = 40% r3,4 = 50%

an Ivo th E , C n), J ate

(1 + r1,4 ) = (1 + 30%) · (1 + 40%) · (1 + 50%) = (1 + 173%)

(1 + r1,2 ) · (1 + r2,3 ) · (1 + r3,4 ) = (1 + r1,4 )

2010 2011 2012 2013 2014 2015

(4 elch itio por ly 2 nan Ivo th E h, C n),

15.0% 2.1% 16.0% 32.3% 13.7% 1.3%

at 201 nce o W Ed Cor , Ju e Fi 17.

Annualized Rates of Return

Per-unit standard measures

are statistics that are

rate of accumulation of distance per unit of time.

A per-unit standard for

rates of return: annualized

an Ivo th E , C n), J ate

Return to our example: You

want to annualize our

three-year total holding

To nd the t-year annualized

interest rate, take the t-th

with a holding rate of return of 173%. To find r3 , solve the equation

root of the total return (t is

or, for short,

(4 elch itio por ly 2 nan Ivo th E h, C n),

1.3976 · 1.3976 · 1.3976 ≈ (1 + 173%)

at 201 nce o W Ed Cor , Ju e Fi 17.

return. Almost all rates are quoted in annualized terms instead.

return, called rt , by taking the t-th root:

1 + r0,t = (1 + r0,t )1/t

Translating long-term dollar

returns into annualized

value. Thus, the total 30-year holding rate of return is

(1 + r30 )30 = (1 + 140%)

Solve this by taking the 30th root,

return of 2.01%/2 = 1.005%. The difference is the interest on the interest.

(4 elch itio por ly 2 nan Ivo th E h, C n),

what you earned per year.

(1 + 100%) · (1 – 50%) = 1 + 0% ⇒ r0,2 = 0%

(1 + r0,1 ) · (1 + r1,2 ) = (1 + r0,2 )

at 201 nce o W Ed Cor , Ju e Fi 17.

(1 + 20%) · (1 – 20%) = (1 – 4%) ⇒ r0,2 = –4%

(1 + r0,1 ) · (1 + r1,2 ) = (1 + r0,2 )

return do tell you that you had a loss:

1 + r2 = (1 + r0,2 ) = 1 – 4% = 1 – 2.02% ⇒ r2 ≈ –2.02%

ä More about how arithmetic returns

are “too high” in normally-distributed

annualizing over different time horizons.

an Ivo th E , C n), J ate