Chicago Booth BUS 35000-81/82/85

Summer 2010
Practice Midterm Exam Questions
Version with Solutions

Note: There are fewer problems in the actual Midterm Exam!

Problem 1. Are the following statements True, False or Ambiguous? Provide a short justi-
fication for your answer. (Your score will primarily be determined by your justification.)

(a) In order for investors to be willing to invest their money for the long run, the yield to
maturity on a 15 year bond must always be greater than the yield to maturity on a 5
year bond.

Solution: False. The yield on a 15 year bond could be lower if investors expect lower
interest rates between year 5 and year 15. This could result in forward rates between
years 5 and 15 that are much lower than the yield to maturity on a 5 year bond. For
example suppose that both bonds are pure discount bonds. Suppose that YTM 5 = 10%
and f6 = f7 = · · · = f15 = 5%. Then the price of a 15 year bond with $100 face value
is:
100
B15 = = 38.12
(1.10)5 (1.05)10
This gives:
 1/15
100
YTM 15 = − 1 = 6.6%
38.12
(b) Consider a coupon bond with a price of $80, and a duration of 5. Then if all interest
rates increase by 1%, the price of the coupon bond will decrease approximately by $5.

Solution: False. The price will fall approximately by 5% of $80, which is $4. To
be more precise, the formula for the percentage change in price (from the definition of
duration) is:
∆P ∆Y
≈ −D × ,
P Y
where Y = 1 + y, and y is the bond yield. Here we assumed that ∆Y /Y = 1%,
therefore the percentage price change should be ∆P/P ≈ −5%. If instead we assume
that ∆Y = 1%, then we have to use the modified duration, which is D∗ = D/(1 + y).
But since we don’t know what the current yield y is, we cannot use this formula.
(Nevertheless, using y = 5% is ok, too.)
(c) If two assets are negatively correlated, then there is a portfolio of the two assets with a
variance of zero.

Solution: False in general. The variance of a portfolio is given by:

σP2 = w2 σ12 + (1 − w)2 σ22 + 2w(1 − w)σ1,2
= w2 σ12 + (1 − w)2 σ22 + 2w(1 − w)ρ1,2 σ1 σ2

1
 where w is the portfolio weight in asset 1.
Only if the two assets are perfectly negatively correlated (i.e., if ρ1,2 = −1) then we
can find a portfolio of two assets with zero standard deviation. Indeed, the variance
becomes

σP2 = w2 σ12 + (1 − w)2 σ22 − 2w(1 − w)σ1 σ2

2
= wσ1 − (1 − w)σ2 .

Then σP > 0 unless wσ1 − (1 − w)σ2 = 0 (which implies w(σ1 + σ2 ) − σ2 = 0). Solve
for w:
σ2
w = .
σ1 + σ2
If ρ1,2 > −1, we can show that σP2 is always larger than zero:

σP2 = w2 σ12 + (1 − w)2 σ22 + 2w(1 − w)ρ1,2 σ1 σ2

> w2 σ12 + (1 − w)2 σ22 − 2w(1 − w)σ1 σ2

2
= wσ1 − (1 − w)σ2
≥ 0.

Since the second inquality is strict (it comes from ρ1,2 > −1), it follows that σP > 0.

(d) If one introduces one more asset in the investment opportunity set, then the Sharpe ratio
of the new mean–variance efficient (MVE) portfolio is at least as high as the Sharpe
ratio of the old MVE portfolio.

Solution: True. After introducing one more asset, the efficient frontier always moves
to the left. This means that the tangent line to the efficient frontier also moves to the
left, i.e. its slope (which is the Sharpe ratio) will be the same or higher.

(e) Because of the potential for diversification if two stocks are not perfectly correlated, a
portfolio that invests 50% in each stock must be less risky that a portfolio invested just
in one of the stocks.

Solution: False in general. Although there is a benefit to diversification, if one of the

stocks is very risky then the portfolio could be riskier than an investment in the low
risk stock. For example suppose that σ1 = 0.1, σ2 = 0.3 and ρ1,2 = 0. The variance of
the 50-50 portfolio is:

σP2 = 0.52 × 0.12 + 0.52 × 0.32 + 2 × 0.5 × 0.5 × 0 × 0.1 × 0.3

= 0.025
√
Hence σP = 0.0250 = 0.1581. This has higher risk than investing entirely in stock #1.

(f) The CEO of XYZ corporation suddenly announced that he would step down due to
health reasons. Since this represents purely idiosyncratic risk, according to CAPM it
should not be priced. Therefore the price of XYZ should not change after the announce-
ment.

2
 Solution: False. CAPM only implies that XYZ’s expected return (or discount rate)
should not change after the announcement (assume that beta stays constant). Never-
theless, the price according to the PV formula depends not just on the discount rate,
but also on the expected cash flows. If for example the market expects that XYZ’s cash
flows will suffer from the CEO departure, then the price should fall.

Problem 2. You are analyzing two stocks: Ford (F) and General Motors (GM). Depending
on how the car industry does during the next year, you propose the following return estimates
under three different scenarios:
Scenario 1 2 3
Probability 1/3 1/3 1/3
r̃F 0.30 0.10 -0.10
r̃GM 0.10 0.30 -0.10
Compute the expected return, return standard deviation, and return correlation for Ford and
GM.
Solution: We compute
1 1 1
EF = × 30% + × 10% + × (−10%) = 10%
3 3 3
1 1 1
EGM = × 10% + × 30% + × (−10%) = 10%
3 3 3
1 1 1
σF2 = × (0.3 − 0.1)2 + × (0.1 − 0.1)2 + × (−0.1 − 0.1)2 = 0.0267
3 3 3
2 1 1 1
σGM = × (0.1 − 0.1)2 + × (0.3 − 0.1)2 + × (−0.1 − 0.1)2 = 0.0267
3 3 3
√
σF = σGM = 0.0267 = 0.1633 = 16.33%.
To compute the correlation between F and GM, compute the covariance first:
1 1 1
σF,GM = × (0.3 − 0.1)(0.1 − 0.1) + × (0.1 − 0.1)(0.3 − 0.1) + × (−0.1 − 0.1)(−0.1 − 0.1)
3 3 3
= 0.0133
σF,GM 0.0133
ρF,GM = = = 0.5 = 50%.
σF σGM 0.1633 × 0.1633

Problem 3. You are in charge of the bond trading and forward loan department of a large
investment bank. Market prices for 1, 2 and 3 year pure discount bonds (zero coupon bonds)
with face value of $100 are displayed on your computer terminal as follows:
Years to Maturity Price
1 98
2 96
3 90

3
 (a) A new summer intern from Harvard has just told you that he thinks that 3 year treasury
notes with annual coupons of $30 and face value of $1,000 are trading for $1,000. Would
you ask the intern to recheck the price of this coupon bond? If so, why?

Solution: This bond should trade at the price:

98 96 90
Price = × 30 + × 30 + × 1, 030
100 100 100
= 985.20

Since this is different than the price the intern thinks it should be, I would ask him to
recheck his numbers.

(b) A customer approaches you looking for a quote on a loan of $20 million dollars to be
received by the customer one year from now. The customer will repay the loan two
years from now. What forward interest rate would you quote for your customer?

Solution:
B1 98
f2 = −1= − 1 = 2.0833%
B2 96
I would quote a rate of 2.0833%.

(c) Suppose that your customer is willing to enter into the loan agreement of part (b). How
would you structure your holdings of zero coupon bonds so that you can exactly match
the future cash flows of this loan?

Solution:

• Since we want to get $20M in year one, I would buy 20M 100
= 200,000 1-year zero
coupon bonds (with face value $100) at a price of 200, 000 × 98 = 19, 600, 000.
• Finance this by selling 19, 600, 000/96 = 204, 167 2-year zero coupon bonds. Notice
this is the same as 20M (1+0.020833)
100
= 204, 167 bonds.

Problem 4. Suppose that the yield to maturity on all bonds is 5% (in other words, the
yield curve is flat at 5%). You have an obligation to pay $10M in 4 years. You have only
two investment opportunities: a 1 year Treasury strip with face value of $100 and a 10 year
Treasury strip with face value of $100.

(a) How much money do you have to invest today in the bond market to entirely fund your
obligation?

Solution: I would have to invest the present value of this obligation, which is:
10M
= 8, 227, 024.75
1.054

4
 (b) How would you structure your holdings of the 1-year and 10-year Treasury strips so
that you are protected against the risk of interest rate fluctuations? How many 1-year
strips and how many 10-year strips would you buy?

Solution: The duration of my obligation is 4. I would like a bond portfolio with this
same duration. The duration of a strip equals to its maturity. If x is the percentage of
my investment in 1-year strips and 1 − x in 10-year strips, then I need:
x + 10(1 − x) = 4
This implies that x = 10−4
10−1
= 23 = 0.6667. Hence 66.67% of my money should be in the
1-year strip and 33.33% of my money should be in the 10-year strip. I should invest
2
3
×8, 227, 024.75 = 5, 484, 683.17 in the 1-year strip and 31 ×8, 227, 024.75 = 2, 742, 341.58
in the 10-year strip.
The price of the 1-year strip is:
100
B1 = = 95.2381
1.05
The price of the 10-year strip is:
100
B10 = = 61.3913
1.0510
Hence I should buy 5, 484, 683.17/95.2381 = 57, 589 1-year strips and 2, 742, 341.58/61.3913 =
44, 670 10-year strips.

Problem 5. Suppose that you are considering investing in two stocks. After analyzing the
two stocks you think that there are two possible states for the economy over the next year:
“Good” and “Bad.” Each state is equally likely (probability 0.5). The returns of the two
securities in each state are as follows:
State Return to Stock #1 Return to Stock #2
Good 30% 5%
Bad 10% 10%
(a) What is the expected return and standard deviation of each stock return?

Solution: The expected return on stock #1 is:

E1 = 0.5 × 0.30 + 0.5 × 0.10 = 0.20 = 20%
and on stock #2 it is:
E2 = 0.5 × 0.05 + 0.5 × 0.10 = 0.075 = 7.5%.
The variances of the stock returns are:
σ12 = 0.5 × (0.3 − 0.2)2 + 0.5 × (0.10 − 0.2)2
= 0.01
2
σ2 = 0.5 × (0.05 − 0.075)2 + 0.5 × (0.10 − 0.075)2
= 0.000625

5
 The standard deviations of the stock returns are:
√
σ1 = 0.01 = 0.1 = 10%
√
σ2 = 0.000625 = 0.025 = 2.5%.

(b) What is the covariance and the correlation between the two stock returns?
Solution: The covariance between the two stocks is:
σ1,2 = 0.5 × (0.30 − 0.20)(0.05 − 0.075) + 0.5 × (0.10 − 0.20)(0.10 − 0.075) = −0.0025
The correlation between the two stocks is:
σ1,2 −0.0025
ρ1,2 = = = −1
σ1 σ2 0.1 × 0.025
(c) Draw a picture to illustrate the tradeoff between risk and return that is available by
investing in these two stocks.
Solution: We know that when two assets are perfectly negatively correlated, i.e. when
ρ1,2 = −1, one can achieve perfect diversification. This means that the investment
opportunity set (which is a broken line) contains a portfolio P with zero standard
deviation. But how do we find it? Suppose P is invested w1 in Stock #1 and w2 = 1−w1
in Stock #2. Then we saw in the lecture notes that
σP = |w1 σ1 − w2 σ2 |
(It has to be the absolute value of the difference, since the standard deviation is always
positive.) So we need to solve
σ2 0.025
w1 σ1 − (1 − w1 )σ2 = 0 =⇒ w1 = = = 20%.
σ1 + σ2 0.025 + 0.1
So w1 = 20% and w2 = 80%. Then EP = w1 E1 + w2 E2 = 0.2 × 20% + 0.8 × 7.5% = 10%.
The investment opportunity set is then given in the following picture:
Investment Opportunity Set with 2 Assets
0.25

0.2 Stock #1
Expected Return

0.15

0.1

Stock #2

0.05

0
0 0.02 0.04 0.06 0.08 0.1 0.12
Standard Deviation

6
 (d) Suppose that a risk-free investment of 5% is also available. Does this present a profit
opportunity to you? Why or why not?
Solution: It is possible to create a risk-free return in the two stocks by holding the
portfolio P from before: 20% in stock #1 and 80% in stock #2. This has a return of
EP = 10%. As a result, there is a significant arbitrage opportunity: I would borrow at
the risk-free rate and invest in P .

Problem 6. You work in the bond trading group of a large investment bank. You are told
that 1, 2 and 3 year treasury strips (zero coupon bonds) are currently trading at the following
implied yields to maturity
Time to Maturity (j) YTM j
1 5%
2 4%
3 8%
(a) The commercial loan department of your firm has asked for your help in constructing
a forward loan for one of their customers. This customer would like to enter into a
contract to borrow $100 million from your firm a year from now to be repaid 2 years
from now. The customer is willing to pay 7% interest on the loan. Should your firm
enter into this contract? If so, how would you structure your holdings of Treasury strips
so that your firm can exactly match the cash flows required by the loan.
Solution: Assuming that the face value of Treasury strips is $100, the prices of the 1-,
2- and 3-year strips are given by:
100 100 100
B1 = = 95.24, B2 = = 92.46, B3 = = 79.38
1.05 1.042 1.083
The implied forward rate between years 1 and 2 is given by:
B1 (1 + YTM 2 )2
f2 = −1 = − 1 = 3.01%.
B2 1 + YTM 1
Since the customer is willing to pay 7% the firm should agree to the contract. To match
the cash flows the firm could do the following:

i. Buy 1M 1-year strips yielding $100M in year 1 to be delivered to the customer.

This costs $1M × B1 = $95.24M .
ii. Finance the purchase of the 1-year strips by selling 1.07M 2-year strips. This
covers the $107M liability in year 2, and produces 1.07M × B2 = $98.93M today.
The profit today from engaging into the contract is $98.93M −$95.24M = $3.69M .

(b) You are told that a 3 year U.S. Treasury note is to be issued today with an annual
coupon of $50 and a face (principal) value of $1,000. What is the highest price that
your group should pay for this bond?
Solution: The most that you should pay for this bond is:
50 50 1050
+ 2
+ = 927.37
1.05 1.04 1.083

7
 (c) Suppose that you purchased the bond in part (b) at the price that you calculated. It
is now one year later and you just received the first coupon payment on the bond. At
this time the yields to maturity on 1, 2 and 3 year Treasury strips are:

Time to Maturity (j) YTM j

1 6%
2 7%
3 8%

If you were to sell the bond now, what rate of return would you realize on your investment
in the bond?

Solution: The price of the bond should now be:

50 1050
+ = 964.28 .
1.06 1.072
The realized rate of return is:
964.28 + 50
− 1 = 9%
927.37

Problem 7. You are an investment manager for Lemon County. You are given the following
information about the yields of U.S. Treasury strips (which are zero-coupon, or pure discount,
bonds which pay $100 at maturity): YTM 1 = 3%, YTM 2 = 4%, YTM 29 = 6.1%, and
YTM 30 = 6.0%, where YTM j is the yield-to-maturity on a j-year strip. Assume that all cash
flows are riskless and we can borrow and lend at the stated rates. You buy 50, 000 30-year
strips and partially finance these by issuing 5, 000 2-year strips (i.e. you are borrowing money
by promising to pay $500, 000 = 5, 000 × $100 in two years).

(a) How much of Lemon County’s money are you investing (i.e. what is the net investment)?

Solution: Cost of 50,000 30-year strips:

100
50, 000 × = 870, 550.65
1.0630
Income from issuing 5,000 2-year strips:
100
5, 000 × = 462, 278.11
1.042
Therefore the net value is

870, 550.65 − 462, 278.11 = 408, 272.54.

8
 (b) Suppose that one second after you make the above investments, the Federal Reserve
Board announces that it is taking steps to raise interest rates. Assume that the yields
on all of the strips immediately increase by one percentage point (i.e. YTM 1 = 4%,
YTM 2 = 5%, YTM 29 = 7.1%, and YTM 30 = 7.0%). By how much (in dollars and in
percentage terms) does the value of the county’s investment change due to the unex-
pected announcement?
Solution: The net value right after the change is
100 100
50, 000 × 30
− 5, 000 × = 656, 835.59 − 453, 514.74 = 203, 320.85.
1.07 1.052
Therefore the dollar change in value is
203, 320.85 − 408, 272.54 = −204.951.69,
and the percentage change in value is
−204, 951.69/408, 272.54 = −0.502 = −50.2%.

(c) After the announcement, what are the forward rates of interest f2 and f30 ?
Solution:
(1 + YTM 2 )2 1.052
f2 = −1 = − 1 = 6.01%
1 + YTM 1 1.04
(1 + YTM 30 )30 1.0730
f30 = −1 = − 1 = 4.14%.
(1 + YTM 29 )29 1.07129

Problem 8. You have been hired by a small pension fund to help them design a bond portfolio
to fund a $10 million obligation that will come due in 4 years. The managers of the fund
would like to use a 2 year zero coupon bond along with an 8 year zero coupon bond to fund
this obligation. Suppose that the yield curve is flat so that the yields to maturity on all zero
coupon bonds are 5%.
(a) Design a portfolio of the two bonds that will protect the pension fund from fluctuations
in interest rates.
Solution: The present value of the obligation is:
10, 000, 000
PV = = 8, 227, 024.74.
1.054
Therefore the fund needs to invest $8, 227, 024.74.
The duration of the obligation is 4. The duration of the 2-year bond is 2 and the
duration of the 8-year bond is 8. Let x be the percentage of the investment in the
2-year. Then:
4 = 2x + 8(1 − x).
This implies that x = 2/3. Hence the fund should invest 32 ×8, 227, 024.74 = 5, 484, 683.16
in the 2-year zero coupon bond. And 13 × 8, 227, 024.74 = 2, 742, 341.58 in the 8-year
zero coupon bond.

9
 (b) Suppose that right after you create this portfolio, the yield curve shifts to 6% at all
maturities. Calculate what you expect the future value of the investment in the two
bonds to be in year 4. Do you meet the obligation of the fund? Explain any difference.

Solution: The above investments result in a cash flow of 5, 484, 683.16 × 1.052 =
6, 046, 863.18 in year two and a cash flow of 2, 742, 341.58 × 1.058 = 4, 051, 687.50 in
year eight. When interest rates are 6% the future value at year 4 of reinvesting the cash
received in year 2 is:

6, 046, 863.18 × 1.062 = 6, 794, 255.47

The value of the cash flow from the 8-year zero when sold in year 4 when interest rates
are 6% is:
4, 051, 687.50
= 3, 209, 315.99
1.064
The total received would be $10, 003, 571.47 > 10, 000, 000. It is larger because this
strategy has convexity. When interest rates rise the duration of the portfolio shifts
towards the 2-year bond. As a result the portfolio is exposed to greater reinvestment
risk. Since interest rates have risen this is an additional benefit to the portfolio.

Problem 9. Suppose you can invest in a risk-free T-bill with rate of return rf = 5%, and
in two risky assets: A = a fixed income fund, and B = an equity fund, which have expected
return & standard deviation:
Asset E(r̃) σ
A 9% 16%
B 14% 20%

The correlation between the real estate and stock funds is ρAB = 0.4. Find the MVE (tan-
gency) portfolio T and the GMV (Global Minimum Variance) for the two risky assets.
Solution: Compute the covariance matrix Ω:

σA,A = σA2 = 0.162 = 0.0256

σA,B = ρA,B σA σB = 0.4 × 0.16 × 0.20 = 0.0128
σB,A = σA,B = 0.0128
σB,B = σB2 = 0.202 = 0.0400.

Compute also the excess returns

EAe = 9% − 5% = 4%
EBe = 14% − 5% = 9%

We get    
0.0256 0.0128 e 0.04
Ω= , E =
0.0128 0.0400 0.09

10
w = wMVE solves (after normalization) Ωw = Ee :

0.0256 wA + 0.0128 wB = 0.04

0.0128 wA + 0.0400 wB = 0.09.

Multiply the first equation by −1 and add it to the second equation multiplied by 2 (to
eliminate wA ). We get
2 × 0.09 − 0.04 0.04 − 0.0128 × wB
wB = = 2.0833, wA = = 0.5208
2 × 0.0400 − 0.0128 0.0256
Normalize by the sum wA + wB = 2.6042 and get
wA wB
wA = = 0.2, wB = = 0.8
2.6042 2.6042
So  
0.2
wMVE =
0.8
w = wGMV solves (after normalization) Ωw = 1:

0.0256 wA + 0.0128 wB = 1
0.0128 wA + 0.0400 wB = 1.

By the same method, this yields wA = 31.62, wB = 14.88 and after normalization wA =
0.68, wB = 0.32. So  
0.68
wGMV =
0.32
Alternatively, for the MVE portfolio you can use the formula in the notes:

EAe σB2 − EBe σA,B

wA =
EAe σB2 + EBe σA2 − (EAe + EBe )σA,B
.04 × .04 − .09 × .0128
=
.04 × .04 + .09 × .0256 − (.04 + .09) × .0128
= 0.2.

Therefore wB = 1 − wA = 0.8. This means that T invests 20% in A and 80% in B.

Problem 10. You are a mean–variance investor, and you decide to choose a portfolio with
the available assets. You are allowed to invest only in an asset that has a risk-free rate of
return rf = 5%, and in two risky assets, GE and KO, which have the following expected
return & standard deviation:
Asset E(r̃) σ
GE 10% 15%
KO 12% 19%

The correlation between the two stocks is ρGE,KO = 0.4.

11
(a) Denote by T the portfolio formed with GE and KO that you would hold. Does it depend
on your risk aversion?

Solution: Your risky-only portfolio does not depend on your risk aversion. It must be
the tangency (MVE) portfolio. You can use the formula
e 2 e
E(r̃GE )σKO − E(r̃KO ) cov(r̃GE , r̃KO )
wGE = e 2 e 2 e e
.
E(r̃GE )σKO + E(r̃KO )σGE − [E(r̃GE ) + E(r̃KO )] cov(r̃GE , r̃KO )
To compute this, you need the expected excess returns E(r̃e ) = E(r̃) − rf , and variances
σ2:
Asset E(r̃e ) σ2
GE 5% 0.0225
KO 7% 0.0361

and also the covariance

cov(r̃GE , r̃KO ) = ρGE,KO σGE σKO = .4 × .15 × .19 = .0114.
Plug in these values:
.05 × .0361 − .07 × .0114
wGE = = .5 = 50%.
.05 × .0361 + .07 × .0225 − (.05 + .07) × .0114
This means that T invests 50% in GE and 50% in KO.
(b) Knowing that the weights are as in part (a), find the expected return and standard
deviation of T .

Solution: Recall: If X and Y are random variables with means µX , µY , standard

deviations σX , σY and correlation ρXY , then a linear transformation of X and Y
Z = αX + βY (1)
will have mean and standard deviation
µZ = αµX + βµY (2)
q
2
σZ = α2 σX + β 2 σY2 + 2αβ cov(X, Y ) (3)

where
cov(X, Y ) = ρXY σX σY .

In this case, the MVE portfolio is a linear transformation of assets GE and KO

T = .5GE + .5KO (4)
where α = .5 and β = .5.
Plugging into the above formula,
µT = .5 × .10 + .5 × .12 = .11 = 11% and
√
σT = .52 × .152 + .52 × .192 + 2 × .5 × .5 × .4 × .15 × .19 = .143 = 14.3%

12
 (c) Find the Sharpe ratio of the tangency portfolio, and show that it is higher than the
Sharpe ratio for both GE and KO.

Solution: The Sharpe ratio of T is

ET − rf .11 − .05
SRT = = = .42
σT .143
while the Sharpe ratios of GE and KO are:
.10 − .05
SRGE = = .33
.15
.12 − .05
SRGE = = .37
.19
Indeed, SRT = .42 is larger than both .33 and .37.
(d) Suppose your utility function is quadratic:

U (E, σ) = E − 12 A σ 2

and your risk aversion coefficient is A = 7. Find your optimal portfolio formed with
the risk-free asset, GE, and KO. What percentage of your wealth will you hold in each
asset?

Solution: Modern Portfolio Theory says that you should only invest in a combination
of the risk-free asset and the MVE portfolio T . The weight that you should put on T is
given by the following formula:
1 µT − r f 1 .11 − .5
wT = 2
= = 0.42.
A σT 7 .1432
So you should hold 42% of your wealth in T , and 58% in the risk-free asset. But T is
formed by GE and KO with equal weights, so you will hold 21% in GE, 21% in KO,
and 58% in the risk-free asset.

Problem 11. You have $1 million currently invested entirely in mutual fund A. You are
considering switching into a combination of T-bills and mutual fund B for the next year.
Mutual fund B is invested 50% in the stock of ABC and 50% in the stock of XYZ. A 1-year
T-bill with face value $10,000 is currently selling for $9,523. You have come up with the
following assessments of the return to mutual fund A along with the returns to ABC and
XYZ for the next year.

E(r̃A ) = 0.10 E(r̃ABC ) = 0.10 E(r̃XY Z ) = 0.18

σA = 0.20 σABC = 0.20 σXY Z = 0.30

You have also determined that ρABC,XY Z = 0.5.

Using just the T-bill and mutual fund B show how you can construct an investment of
your $1 million such that the portfolio is better than your current investment in mutual fund
A. You must show clearly why the new portfolio is better and indicate the dollar investment
in T-bills and in mutual fund B.

13
Solution: To solve this problem we need the expected return and standard deviation of the
return to mutual fund B. The expected return is given by:

E(r̃B ) = 0.5 × E(r̃ABC ) + 0.5 × E(r̃XY Z )

= 0.5 × 10% + 0.5 × 18%
= 14%

The variance of the return to mutual fund B is given by:

σB2 = 0.52 × σABC

2
+ 0.52 × σXY
2
Z + 2 × 0.5 × 0.5 × ρABC,XY Z × σABC × σXY Z
= 0.52 × 0.22 + 0.52 × 0.32 + 2 × 0.5 × 0.5 × 0.5 × 0.2 × 0.3
= 0.0475

Hence σB = 0.218.
We also need the 1-year risk-free rate. This is given by the price of the one year T-Bill:
10, 000
1 + rf = = 1.05
9523
Hence rf = 5%.
Let’s find a portfolio with the same return as mutual fund A. Let w be the weight in this
portfolio invested in mutual fund B. Then

(1 − w) × 5% + w × 14% = 10%

This implies that w = 5/9. The standard deviation of this portfolio’s return is σP = 5/9 ×
0.218 = 12.1%. The portfolio has the same expected return as mutual A, but lower standard
deviation.
Alternatively we could find a portfolio with the same return standard deviation as that of
mutual fund A. In this case we set:

w × 0.218 = 0.20

This implies that w = 0.917. The expected return of this portfolio is

(1 − 0.917) × 5% + 0.917 × 14% = 13.25%.

The expected return on this portfolio is larger than that of mutual fund A.

Problem 12. A portfolio consists of the following 3 stocks, whose performance depends on
the economic environment. Assume the bad economic outcome is twice as likely as the good
one.
Investment in $(mill.) Good economy Bad economy
Stock 1 500 13% −20%
Stock 2 1,250 6% 3%
Stock 3 250 −7% 2%

(a) Calculate the expected return of the portfolio.

Solution:

14
 Investment in $(mill.) Good economy Bad economy
Stock 1 500 13% −20%
Stock 2 1,250 6% 3%
Stock 3 250 −7% 2%
r̃P 2,000 .25(.13) + .625(.06) + .125(-.07) .25(-.20) + .625(.03) + .125(.02)
= 0.06125 = -0.02875

So the expected return of the portfolio is

EP = 1/3(0.06125) + 2/3(−0.02875) = 0.00125.

(b) Calculate the variance of the portfolio.

Solution: The variance of the portfolio is
σP2 = 1/3(0.06125 − 0.00125)2 + 2/3(−0.02875 − 0.00125)2 = 0.0018.

(c) What if we add to the portfolio $1,000 million of stock 4, which has a mean return of
4%, has a variance of 0.02, and is uncorrelated with the above portfolio? How will this
change the expected return and variance of the total investment?
Solution: Call this new portfolio Q. The new expected return and variance are
1, 000 2, 000
EQ = (0.04) + (0.00125) = 0.0142,
3, 000 3, 000
 2  2
1 2
σQ2 = (0.02) + (0.0018) = 0.00302.
3 3

Problem 13. The expected return of the S&P 500 index, which turns out to be the tangency
portfolio of stocks in the economy, is 16%, with a standard deviation of return of 25% per
year. The expected return of Microsoft is unknown, but it has a standard deviation of 20%
per year, and a covariance with the S&P 500 index of 0.10. Assume that the risk-free rate is
6% per year.
(a) Compute Microsoft’s beta and expected return.
Solution:
σMSFT,M 0.10
βMSFT = 2
= = 1.60,
σM (0.25)2
EMSFT = 0.06 + 1.60(0.16 − 0.06) = 0.22.

(b) If Intel has half the expected return of Microsoft, then what is Intel’s beta?
Solution:
0.22
EINTC =
= 0.11,
2
CAPM =⇒ 0.11 = 0.06 + βINTC (0.16 − 0.06),
βINTC = 0.50.

15
 (c) What is the beta and expected return of the following portfolio?

• 25% in Microsoft
• 10% in Intel
• 75% in the S&P 500 index
• −20% in GM (where βGM = 0.80)
• 10% in the risk-free asset

Solution:

βP = .25 × 1.60 + .10 × .50 + .75 × 1 − .20 × .80 + .10 × 0 = 1.04,

EP = 0.06 + 1.04(0.16 − 0.06) = 0.164.

(d) Is the following statement True, False, or It Depends?

If the risk-free rate drops to 5%, but the expected return on the S&P 500
index remains the same, the new expected return on Microsoft increases by
0.6%.

In one or two sentences explain or show why this is true or false or why it depends.

Solution: It depends, because once the risk-free asset changes, the S&P 500 may no
longer be the tangency portfolio. Therefore, we don’t know what the new expected
return equation looks like. Remember, the S&P 500 is simply some portfolio, which
happens to be the tangency portfolio given a risk-free rate of 6%, but it does not
necessarily represent the market portfolio in a CAPM context.

Problem 14. Suppose CAPM is true, and that the market portfolio M contains only two
risky assets: IBM and GE. The expected returns of IBM and GE are 10% and 9%, respectively.
Furthermore, the variance-covariance matrix for these two stocks is given by the following:
IBM GE
IBM 0.004 0.001
GE 0.001 0.002
The risk-free rate is 4%.

(a) Use the CAPM equation to show that, if M is the market portfolio, and i and j are any
two risky assets in the market portfolio, then the following equation holds:
Ei − rf Ej − rf
=
σi,M σj,M

Solution: CAPM says that expected returns are linear in covariance:

σi,M EM − rf
Ei − rf = βi (EM − rf ) = 2
(EM
− r f ) = σi,M
× 2
,
σM σM

16
 therefore
Ei − rf E − rf
= M 2 ,
σi,M σM
which does not depend on the particular asset i.

(b) Find the weights of the market portfolio. (Hint: Denote by w1 the weight of IBM in the
market portfolio, and by w2 = 1 − w1 the weight on GE. Then use the equation in (a)
to find an equation that w1 satisfies.)

Solution: Let w1 be the weight of IBM in the market portfolio, and w2 = 1 − w1 the
weight of GE. Then the return of the market portfolio can be written: rM = w1 r1 + (1 −
w1 )r2 , where r1 = rGE and r2 = rIBM . Note that the covariance of asset i with the market
is σi,M = cov(ri , rM ) = cov(ri , w1 r1 + (1 − w1 )r2 ) = w1 cov(ri , r1 ) + (1 − w1 ) cov(ri , r2 ) =
w1 σi,1 +(1−w2 )σi,2 . For example, σ1,M = w1 σ1,1 +(1−w2 )σ1,2 = w1 ×.004+(1−w1 )×.001.
We used the fact that σ1,1 = .004 and σ1,2 = .001 are the elements on the first row of
the given covariance matrix. Similarly, we compute σ2,M = w1 σ2,1 + (1 − w2 )σ2,2 =
w1 × .001 + (1 − w1 ) × .002.
Now, using the equation in (a), we obtain
.10 − .04 .09 − .04
=
.004w1 + .001(1 − w1 ) .001w1 + .002(1 − w1 )

Solving for w1 , we get

w1 = 1/3,
w2 = 2/3.

(c) Suppose a highly risk averse investor hires you to design an asset allocation plan, but
tells you that the tangency portfolio is “too risky” for his/her tastes. What would you
tell this person? Specifically, if this investor informs you he/she would like to be have
a wealth percentage w invested in the risk-free asset, what asset allocation plan would
you recommend? (Be specific, and describe the weights you would place on the risk-free
asset, IBM, and GE.)

Solution: Every investor’s portfolio must be a combination of the risk-free asset and
the market:

w in rf & (1 − w) in M
1 2
⇒ (1 − w) in IBM & (1 − w) in GE
3 3
If you are risk averse, just increase w!

(d) Is the following statement True, False, or It Depends?

A portfolio with the same expected return as IBM has the same correlation
with the market portfolio.

17
 In one or two sentences explain or show why this is true or false or why it depends.
Solution: False.
Assets with the same expected return have the same covariance with the market port-
folio, but not necessarily the same correlation.
σi,M = σj,M
=⇒ ρiM σi σM = ρjM σj σM .
Therefore, it depends on whether σi = σj (i.e. whether the two assets have the same
standard deviation).

Problem 15. Your firm is evaluating a new risky capital investment. After careful consid-
eration, you decide there are two possible outcomes or states of the world relevant to your
project. State 1 represents a mild recovery, and state 2 represents a strong recovery from
the recent downturn in the market. You estimate the probabilities of these alternative future
states of the world to be 0.7 and 0.3, respectively. You also estimate that the market return
will be 10% in the mild recovery state and 30% in the strong recovery case. The project
itself has estimated returns of zero in the mild recovery case and 40% in the strong recovery
scenario. Assume that the risk-free rate is 10%.
(a) What are the expected returns of the market and the project, respectively?
Solution: The expected returns of the project P and the market are
EP = .7(0) + .3(.40) = 0.12
EM = .7(.10) + .3(.30) = 0.16

(b) What is the covariance between your project and the market?
Solution: The covariance between the project and the market is
σP,M = .7(0 − .12)(.10 − .16) + .3(.40 − .12)(.30 − .16) = 0.0168

(c) Under the CAPM, what is your expected return on the project?
Solution: Under CAPM the expected return on P is EP = .10 + βP (.16 − .10). To
compute βP , compute first market variance
2
σM = .7(.10 − .16)2 + .3(.30 − .16)2 = 0.0084.
Then
σP,M 0.0168
βP = 2
= = 2.
σM 0.0084
This implies
EP = .10 + 2(.16 − .10) = 0.22.
(d) Should the project be accepted? Why or Why not? (Be brief.)
Solution: No.
Since 12% (the actual expected return of the project) < 22% (the CAPM expected
return), the project does not offer a high enough expected return to compensate for its
risk.

18