CHAPTER 7

Optimal Risky Portfolios

INVESTMENTS | BODIE, KANE, MARCUS

McGraw-Hill/Irwin Copyright © 2011 by The McGraw-Hill Companies, Inc. All rights reserved.
 The Investment Decision
Top-down process with 3 steps:
1. Capital allocation between the risky
portfolio and risk-free asset
2. Asset allocation across broad asset
classes
3. Security selection of individual assets
within each asset class

INVESTMENTS | BODIE, KANE, MARCUS 7-2

 Diversification and Portfolio Risk
• Market risk
– Risk attributable to market-wide risk
sources, and remains even after extensive
diversification
– aka Systemic or non-diversifiable

• Firm-specific risk
– Risk that can be eliminated by diversification
– aka diversifiable or non-systemic

INVESTMENTS | BODIE, KANE, MARCUS 7-3

 Fig. 7.1 Portfolio Risk as a Function of
the Number of Stocks in the Portfolio

Panel A: All risk is firm specific. Panel B: Some risk is systematic or marketwide.

INVESTMENTS | BODIE, KANE, MARCUS 1-4

Figure 7.2 Portfolio Diversification

INVESTMENTS | BODIE, KANE, MARCUS 1-5

 Covariance and Correlation
• Portfolio risk depends on the correlation
between the returns of the assets in the
portfolio

• Covariance and the correlation coefficient

provide a measure of the way returns of two
assets vary

INVESTMENTS | BODIE, KANE, MARCUS 7-6

A Two-Security Portfolio: Return
𝑤𝐷 = Bond weight
𝑤𝐸 = Equity weight

𝑟𝐷 = Bond return
𝑟𝐸 = Equity return

𝑟𝑃 = Portfolio return

E (rp )  wD E (rD )  wE E (rE )

INVESTMENTS | BODIE, KANE, MARCUS 7-7
A Two-Security Portfolio: Risk

  w   w   2wD wE CovrD , rE 
2
p
2
D
2
D
2
E
2
E

 = Variance of Security D
2
D

 2
E = Variance of Security E

CovrD , rE  = Covariance of returns for

Security D and Security E

INVESTMENTS | BODIE, KANE, MARCUS 7-8

 Two-Security Portfolio: Risk

• Another way to express variance of the

portfolio is to think of Covariances:

  wD wDCovrD , rD  
2
p

 wE wE CovrE , rE  
 2wD wE CovrD , rE 

INVESTMENTS | BODIE, KANE, MARCUS 7-9

 Covariance

CovrD , rE    DE D E
𝜌𝐷,𝐸 = Correlation coefficient of returns

𝜎𝐷 = Standard deviation of returns

for Security D

𝜎𝐸 = Standard deviation of returns

for Security E

INVESTMENTS | BODIE, KANE, MARCUS 7-10

 Table 7.2 Computation of Portfolio
Variance From the Covariance Matrix

INVESTMENTS | BODIE, KANE, MARCUS 1-11

 A portfolio of 3 Assets

• You have three assets with weights

w 1, w 2, w 3
• The portfolio return is simply the linear
combination of the returns with same
coefficients:
E (rp )  w1E (r1 )  w2 E (r2 )  w3 E (r3 )
Q. is the portfolio’s variance also the
linear combination of the 3 variances?
INVESTMENTS | BODIE, KANE, MARCUS 1-12
 Bordered Matrix for 3 Assets
Step 1: write the covariance matrix and its weights
w1 w2 w3
w1 Cov(1,1) Cov(1,2) Cov(1,3)

w2 Cov(2,1) Cov(2,2) Cov(2,3)

w3 Cov(3,1) Cov(3,2) Cov(3,3)

INVESTMENTS | BODIE, KANE, MARCUS 7-13

 Bordered Matrix for 3 Assets
Step 2: Symmetry! Cova, b   Covb, a    a ,b
w1 w2 w3
w1  2
1
 1, 2  1,3

w2  1, 2  2
2
 2,3

w3  1,3  2,3  2
3

INVESTMENTS | BODIE, KANE, MARCUS 7-14

 Bordered Matrix for 3 Assets
Step 3: multiply by the weights around the border
w1 w2 w3
w1 w 
2
1
2
1
w1w2  1, 2 w1w3  1,3

w2 w1w2  1, 2 w 2
2  2
2
w2 w3  2,3

w3 w1w3  1,3 w2 w3  2,3 w  2

3
2
3

INVESTMENTS | BODIE, KANE, MARCUS 7-15

 Bordered Matrix for 3 Assets
Step 4: add-up all the pieces

 w  w  w 
2
p
2
1
2
1
2
2
2
2
2
3
2
3

 2 w1w2 1, 2  2 w1w3 1,3  2 w2 w3 2,3

Covariance terms

Remember  a ,b   b ,a   a ,b a b   b ,a b a

INVESTMENTS | BODIE, KANE, MARCUS 1-16
 Bordered Matrix for 3 Assets
All in one step, using correlations this time
w1 w2 w3
w1 w 
2
1
2
1 w1w2 1, 2 1 2 w1w3 1,3 1 3

w1w2 1, 2 1 2 w  w3 w3 2,3 2 3

2 2
w2 2 2

w3 w1w3 1,3 1 3 w3 w3  2,3 2 3 w  2

3
2
3

INVESTMENTS | BODIE, KANE, MARCUS 7-17

 Bordered Matrix for 3 Assets
Add-up all the pieces

 w  w  w 
2
p
2
1
2
1
2
2
2
2
2
3
2
3

 2 w1w2 1, 2 1 2
 2 w1w3 1,3 1 3
 2 w2 w3  2,3 2 3

INVESTMENTS | BODIE, KANE, MARCUS 1-18

Correlation: Possible Values

Range of values for correlation 

−𝟏 ≤ 𝝆 ≤ +1
If  = 1.0, the securities are perfectly
positively correlated
If  = - 1.0, the securities are perfectly
negatively correlated

INVESTMENTS | BODIE, KANE, MARCUS 7-19

 Two-Security Portfolio: Variance
• Remember the variance of a two-asset
portfolio

 w  w 
2
p
2
D
2
D
2
E
2
E

 2 wD wE  DE D E

INVESTMENTS | BODIE, KANE, MARCUS 7-20

 Correlation Coefficients
• When 𝜌𝐷𝐸 = 1, there is no diversification
 P  wE E  wD D
• When 𝜌𝐷𝐸 = −1, a perfect hedge is when:

  w   w   2wD wE D E  0
2
p
2
D
2
D
2
E
2
E

the solution (which also makes 𝑤𝐷 + 𝑤𝐸 = 1) is:

E D
wD  and wE   1  wD
 D  E  D  E
INVESTMENTS | BODIE, KANE, MARCUS 7-21
Fig 7.3 Portfolio Expected Return as a
Function of Investment Proportions

INVESTMENTS | BODIE, KANE, MARCUS 1-22

Fig 7.4 Portfolio Standard Deviation as a
Function of Investment Proportions

INVESTMENTS | BODIE, KANE, MARCUS 1-23

 The Minimum Variance Portfolio

• The minimum variance • If correlation < +1

portfolio is the portfolio
composed of the risky
assets that has the
smallest standard
deviation, the portfolio
with least risk.
• If correlation = -1
the portfolio standard
deviation may be
smaller than that of
either of the individual
component assets.
the standard deviation
of the minimum
variance portfolio is
zero.INVESTMENTS |
BODIE, KANE, MARCUS 7-24
 Fig 7.5 Portfolio Expected Return as a
Function of Standard Deviation

Portfolio
opportunity
set for given
correlation 

INVESTMENTS | BODIE, KANE, MARCUS 1-25

 Correlation Effects
• The amount of possible risk reduction
through diversification depends on the
correlation.
• The risk reduction potential increases as the
correlation approaches -1.
– If 𝜌 = +1.0, no risk reduction is possible.
– If 𝜌 = 0, 𝜎𝑃 may be less than the standard
deviation of either component asset.
– If 𝜌 = -1.0, a riskless hedge is possible.

INVESTMENTS | BODIE, KANE, MARCUS 7-26

Fig 7.6 The Opportunity Set of the Debt and
Equity Funds and Two Feasible CALs

INVESTMENTS | BODIE, KANE, MARCUS 1-27

 The Sharpe Ratio
• Maximize the slope of the CAL for any
possible portfolio, P.
• The objective function is the slope:

E rP   rf
SP 
P
• The slope is also the Sharpe ratio.

INVESTMENTS | BODIE, KANE, MARCUS 7-28

Fig 7.7 The Opportunity Set of Debt and Equity Funds
with the Optimal CAL and the Optimal Risky Portfolio

INVESTMENTS | BODIE, KANE, MARCUS 7-29

Fig 7.8 Determination of the Optimal
Overall Portfolio

INVESTMENTS | BODIE, KANE, MARCUS 1-30

 Markowitz Portfolio Selection Model
• Security Selection
– The first step is to determine the risk-return
opportunities available
– All portfolios that lie on the minimum-variance
frontier from the global minimum-variance
portfolio and upward provide the best risk-return
combinations

INVESTMENTS | BODIE, KANE, MARCUS 7-31

Fig 7.10 The Minimum-Variance
Frontier of Risky Assets

INVESTMENTS | BODIE, KANE, MARCUS 1-32

 Markowitz Portfolio Selection Model
• We now search for the CAL with the highest
reward-to-variability ratio
Q. How do we measure that?

• That means to find that optimal line that

stems from the risk-free point and is tangent
to the efficient frontier

INVESTMENTS | BODIE, KANE, MARCUS 7-33

Fig 7.11 The Efficient Frontier of Risky
Assets with the Optimal CAL

INVESTMENTS | BODIE, KANE, MARCUS 1-34

 Markowitz Portfolio Selection Model
• Everyone invests in P, regardless of
their degree of risk aversion. However:

– More risk averse investors put more in the

risk-free asset.

– Less risk averse investors put more in P.

INVESTMENTS | BODIE, KANE, MARCUS 7-35

Capital Allocation, Separation Property
• The separation property tells us that the
portfolio choice problem may be
separated into two independent tasks:
1. Determination of the optimal risky portfolio
(purely technical / mathematical)
2. Allocation of the complete portfolio to risk
free asset versus the risky portfolio
(depends on personal preference)

INVESTMENTS | BODIE, KANE, MARCUS 7-36

 Fig 7.13 Capital Allocation Lines with
Various Portfolios from the Efficient Set

INVESTMENTS | BODIE, KANE, MARCUS 1-37

 The Power of Diversification
 P2   wi w j Covri , rj 
n n
• Remember:
i 1 j 1
1
• Consider an equally weighted portfolio: wi 
n
• Look at covariance the matrix structure:

• Rewrite covariance as:

Cov ri , rj 
n n n
1 2 11
 P   2 i 
2

i 1 n i 1 j 1 n n
j i

INVESTMENTS | BODIE, KANE, MARCUS 1-38

 The Power of Diversification
• Rearrange:  P2    i2   1 1 Cov ri , rj 
n n n
1 1
n i 1 n i 1 j 1 n n
j i

• Define avg variance and avg covariance as:

n
1
 2    i2
n i 1

Cov ri , rj 
n n
1
Cov  
nn  1 i 1 j 1
j i
 
n  n 1 terms

INVESTMENTS | BODIE, KANE, MARCUS 1-39

 The Power of Diversification

• Then we can rewrite portfolio variance:

Cov ri , rj 
n n n
1 1 2 11
    i  
2
P
ni 1 n
  i 1 j 1 n n
j i
 2  
n  n 1 terms
as:
1 2 n 1
   
2
P Cov
n n
INVESTMENTS | BODIE, KANE, MARCUS 7-40
 The Power of Diversification
Study case where all assets have same
standard deviation and one correlation for all
1 2 n 1
   
2
P  2

n n
Q. What happens for very large n?

  
2
P
2
   
Q. What happens when correlation = 0?
INVESTMENTS | BODIE, KANE, MARCUS 7-41
Table 7.4 Risk Reduction of Equally Weighted Portfolios
in Correlated and Uncorrelated Universes

INVESTMENTS | BODIE, KANE, MARCUS 1-42

 Optimal Portfolios and Non-normal
Returns
• The optimal portfolio approach we just
studied assumes normal returns.
• Fat-tailed distributions can result in extreme
values of Value-at-Risk (VaR) and Expected
Shortfall (ES) and encourage smaller
allocations to the risky portfolio.
• If other portfolios provide sufficiently better
VaR and ES values than the mean-variance
efficient portfolio, we may prefer these when
faced with fat-tailed distributions.
INVESTMENTS | BODIE, KANE, MARCUS 7-43
 Risk Pooling - Insurance Principle
• Risk pooling: merging (adding) uncorrelated,
risky projects as a means to reduce risk.
– increases the scale of the risky investment by
adding additional uncorrelated assets.
• The insurance principle: risk increases less
than proportionally to the number of policies
insured when the policies are uncorrelated
– Sharpe ratio increases

INVESTMENTS | BODIE, KANE, MARCUS 7-44

 Risk Sharing
• As risky assets are added to the portfolio, a
portion of the pool is sold to maintain a risky
portfolio of fixed size.
• Risk sharing combined with risk pooling is the
key to the insurance industry.
• True diversification means spreading a
portfolio of fixed size across many assets, not
merely adding more risky bets to an ever-
growing risky portfolio.

INVESTMENTS | BODIE, KANE, MARCUS 7-45

 Investment for the Long Run

Long Term Strategy Short Term Strategy

• “Invest in the risky • “Invest in the risky
portfolio for 2 years” portfolio for 1 year and
• Long-term strategy is in the risk-free asset for
riskier. the second year”
• Risk can be reduced by
selling some of the
risky assets in year 2.
• “Time diversification” is
not true diversification.
INVESTMENTS | BODIE, KANE, MARCUS 7-46