Page 467

CHAPTER 13
Return, Risk, and the Security Market Line

Iftekkhar/Shutterstock

As of December 2020, Discovery, Inc. (DISCA) posted a return of 26.6%, outperforming the
S&P 500, which finished 2020 with a gain of 16.26% for the year. In hindsight, investing in
DISCA looks better than S&P 500. But can we judge an investment only on the basis of return?
Indeed, one cannot talk about returns in isolation because investment decisions always involve
a trade-off between risk and return. This chapter explores the relationship between risk and
return, and also introduces the important concepts of diversification and asset pricing.

LEARNING OBJECTIVES

After studying this chapter, you should understand:

LO The calculation for expected returns and standard deviations for individual

securities and portfolios.

LO The principle of diversification and the role of correlation.

 LO Systematic and unsystematic risk.

LO Beta as a measure of risk and the security market line.

In our last chapter, we learned some important lessons from capital market history. Most importantly, there is
a reward, on average, for bearing risk. We called this reward a risk premium. The second lesson is that this risk
premium is larger for riskier investments. The principle that higher returns can be earned only by taking greater
risks appeals to our moral sense that we cannot have something for nothing. This chapter explores the
economic and managerial implications of this basic idea.
Thus far, we have concentrated mainly on the return behaviour of a few large portfolios. We need to expand our
consideration to include individual assets and mutual funds. Accordingly, the purpose of this chapter is to
provide the background necessary for learning how the risk premium is determined for such assets.
Page 468

When we examine the risks associated with individual assets, we find two types of risk: systematic and
unsystematic. This distinction is crucial because systematic risks affect almost all assets in the economy, at
least to some degree, while a particular unsystematic risk affects at most a small number of assets. We then
develop the principle of diversification, which shows that highly diversified portfolios tend to have almost no
unsystematic risk.
The principle of diversification has an important implication. To a diversified investor, only systematic risk
matters. It follows that in deciding whether to buy a particular individual asset, a diversified investor is
concerned only with that asset’s systematic risk. This is a key observation, and it allows us to say a great deal
about the risks and returns on individual assets. In particular, it is the basis for a famous relationship between
risk and return called the security market line, or SML. To develop the SML, we introduce the equally famous
beta coefficient, one of the centrepieces of modern finance. Beta and the SML are key concepts because they
supply us with at least part of the answer to the question of how to determine the required return on an
investment.
 LO1 13.1 | Expected Returns and Variances
In the previous chapter, we discussed how to calculate average returns and variances using historical data.
We now begin to discuss how to analyze returns and variances when the information we have concerns future
possible returns and their possibilities.

Expected Return
We start with a straightforward case. Consider a single period of time; say, one year. We have two stocks, L
and U, with the following characteristics. Stock L is expected to have a return of 25% in the coming year,
while Stock U is expected to have a return of 20% for the same period. 1
In a situation like this, if all investors agreed on the expected returns, why would anyone want to hold Stock
U? After all, why invest in one stock when the expectation is that another will do better? Clearly, the answer
must depend on the risk of the two investments. The return on Stock L, although it is expected to be 25%,
could actually be higher or lower.
For example, suppose the economy booms. In this case, we think Stock L would have a 70% return. If the
economy enters a recession, we think the return would be −20%. Thus, we say there are two states of the
economy, meaning that these are the only two possible situations. This setup is oversimplified, of course,
but it allows us to illustrate some key concepts without a lot of computation.
Suppose we think a boom and a recession are equally likely to happen, a 50–50 chance of either.
Table 13.1 illustrates the basic information we have described and some additional information about
Stock U. Notice that Stock U earns 30% if there is a recession and 10% if there is a boom.
Page 469

TABLE 13.1

States of the economy and stock returns

Security Returns if State Occurs

State of the Economy P robability of State of the Economy L U
Recession 0.5 −20% 30%
Boom 0.5 70 10

Obviously, if you buy one of these stocks, say Stock U, what you earn in any particular year depends on
what the economy does during that year. However, suppose the probabilities stay the same through time. If
you hold U for a number of years, you’ll earn 30% about half the time and 10% the other half. In this case,
we say that your expected return on Stock U, E(RU), is 20%:
E(R U ) = (.50 × 30%) + (.50 × 10%) = 20%

In other words, you should expect to earn 20% from this stock, on average.
For Stock L, the probabilities are the same, but the possible returns are different. Here we lose 20% half the
time, and we gain 70% the other half. The expected return on L, E(RL), is thus 25%:
E(R L ) = (.50 × −20%) + (.50 × 70%) = 25%

Table 13.2 illustrates these calculations.

TABLE 13.2

Calculation of expected return

Stock L Stock U
(2) P robability (3) Rate of (5) Rate of
(1) State of of State of Return if State (4) P roduct Return if State (6) P roduct
Economy Economy Occurs (2) × (3) Occurs (2) × (5)
Recession 0.5 −.20 −.10 .30 .15
Boom 0.5 .70 .35 .10 .05
1.0 E(RL) = 25% E(RU) = 20%

In our previous chapter, we defined the risk premium as the difference between the return on a risky
investment and a risk-free investment, and we calculated the historical risk premiums on some different
investments. Using our projected returns, we can calculate the projected or expected risk premium as the
difference between the expected return on a risky investment and the certain return on a risk-free investment.
For example, suppose risk-free investments are currently offering 8%. We say the risk-free rate (which we
label as Rf) is 8%. Given this, what is the projected risk premium on Stock U? On Stock L? Since the
expected return on Stock U, E(RU), is 20%, the projected risk premium is:

Risk premium = Expected return − Risk-f ree rate [13.1]

= E(R U ) − R f

= 20% − 8% = 12%

Page 470

Similarly, the risk premium on Stock L is 25% − 8% = 17%.

In general, the expected return on a security or other asset is simply equal to the sum of the possible returns
multiplied by their probabilities. So, if we have 100 possible returns, we would multiply each one by its
probability and add the results. The result would be the expected return. The risk premium would be the
difference between this expected return and the risk-free rate.
A useful generalized equation for expected return is:

E(R) = ∑ R j × P j
[13.2]
j

where:
R j = value of the jth outcome

P j = associated probability of occurrence

∑ = the sum overall j

j
XAMPLE 13.1
Unequal Probabilities
Look back at Tables 13.1 and 13.2. Suppose you thought that a boom would occur
only 20% of the time instead of 50%. What are the expected returns on Stocks U and L in
this case? If the risk-free rate is 10%, what are the risk premiums?
The first thing to notice is that a recession must occur 80% of the time (1 − .20 = .80)
because there are only two possibilities. With this in mind, Stock U has a 30% return in 80%
of the years and a 10% return in 20% of the years. To calculate the expected return, we
again just multiply the possibilities by the probabilities and add up the results:
E(R U ) = (.80 × 30%) + (.20 × 10%) = 26%

Table 13.3 summarizes the calculations for both stocks. Notice that the expected return
on L is −2%.

TABLE 13.3

Calculation of expected return

Stock L Stock U
(2) P robability (3) Rate of (5) Rate of
(1) State of of State of Return if State (4) P roduct Return if State (6) P roduct
Economy Economy Occurs (2) × (3) Occurs (2) × (5)
Recession .80 −.20 −.16 .30 .24
Boom .20 .70 .14 .10 .02
1.0 E(RL) = −2% E(RU) = 26%

The risk premium for Stock U is 26% − 10% = 16% in this case. The risk premium for
Stock L is negative: −2% − 10% = − 12%. This is a little odd, but, for reasons we discuss
later, it is not impossible.

Page 471

Calculating the Variance

To calculate the variances of the returns on our two stocks, we first determine the squared deviations from
the expected return. We then multiply each possible squared deviation by its probability. We add these, and
the result is the variance. The standard deviation, as always, is the square root of the variance. It is important
to note that later on in the chapter another alternative to calculating variance will be introduced, using the
correlation coefficient.
Generalized equations for variance and standard deviation are:

2
 2

[13.3]
2
σ = ∑ [R j − E(R)] × Pj
j

2
σ = √σ

To illustrate, Stock U has an expected return of E (R ) = 20% . In a given year, it could actually return
U

either 30% or 10%. The possible deviations are thus 30% − 20% = 10% or 10% − 20% = −10% . In
this case, the variance is:
2 2 2
Variance = σ = (.50 × (10%) ) + (.50 × (−10%) ) = .01

The standard deviation is the square root of this:

Standard deviation = σ = √ .01 = .10 = 10%

Table 13.4 summarizes these calculations for both stocks. Notice that Stock L has a much larger
variance.

TABLE 13.4

Calculation of variance

(3) Return Dev iation (4) Squared Return

(1) State of (2) P robability of from Expected Dev iation from (5) P roduct
Economy State of Economy Return Expected Return (2) × (4)
Stock L
Recession 0.5 −.20 − .25 = −.45 (−.45) 2 = .2025 .10125
Boom 0.5 .70 − .25 = .45 (.45) 2 = .2025 .10125
σ2 L = .2025
Stock U
Recession 0.5 .30 − .20 = .10 (.10) 2 = .01 .005
Boom 0.5 .10 − .20 = −.10 (−.10) 2 = .01 .005
σ2 U = .010

When we put the expected return and variability information for our two stocks together, we have:

Stock L Stock U
Expected return, E(R) 25% 20%
Variance, σ 2 .2025 .0100
Standard deviation, σ 45% 10%

Stock L has a higher expected return, but U has less risk. You could get a 70% return on your investment
in L, but you could also lose 20%. Notice that an investment in U always pays at least 10%.
Which of these two stocks should you buy? We can’t really say; it depends on your personal preferences. We
can be reasonably sure that some investors would prefer L to U and some would prefer U to L.
Page 472
 You’ve probably noticed that the way we calculated expected returns and variances here is somewhat
different from the way we did it in the last chapter. The reason is that, in Chapter 12, we were examining
actual historical returns, so we estimated the average return and the variance based on some actual events.
Here, we have projected future returns and their associated probabilities, so this is the information with
which we must work.

XAMPLE 13.2
More Unequal Probabilities
Going back to Example 13.1, what are the variances on the two stocks once we have
unequal probabilities? The standard deviations? We can summarize the needed
calculations as follows:

(1) State of (2) P robability of State of (3) Return Dev iation from (4) Squared Return Dev iation (5) P roduc
Economy Economy Expected Return from Expected Return (2) × (4)
Stock L
Recession .80 −.20 − (−.02) = −.18 .0324 .02592
Boom .20 .70 − (−.02) = .72 .5184 .10368
σ2 L= .1296
Stock U
Recession .80 .30 − .26 = .04 .0016 .00128
Boom .20 .10 − .26 = −.16 .0256 .00512
σ 2 U = .0064

These calculations give a standard deviation for L of σ L = √ .1296 = 36% . The standard
deviation for U is much smaller; σ = √.0064 = 8%.
U

Concept Questions
. How do we calculate the expected return on a security?
. In words, how do we calculate the variance of the expected return?
 LO2 13.2 | Portfolios
Thus far in this chapter, we have concentrated on individual assets considered separately. However, most
investors actually hold a portfolio of assets. All we mean by this is that investors tend to own more than just
a single stock, bond, or other asset. Given that this is so, portfolio return and portfolio risk are of obvious
relevance. Accordingly, we now discuss portfolio expected returns and variances.

Portfolio Weights
There are many equivalent ways of describing a portfolio. The most convenient approach is to list the
percentages of the total portfolio’s value that are invested in each portfolio asset. We call these percentages
the portfolio weights.
Page 473

For example, if we have $50 in one asset and $150 in another, our total portfolio is worth $200. The
percentage of our portfolio in the first asset is $50/$200 = 25% . The percentage of our portfolio in the
second asset is $150/$200 = 75% . Thus, our portfolio weights are .25 and .75 respectively. Notice that
the weights have to add up to 1.00 since all of our money is invested somewhere. 2

Portfolio Expected Returns

Let’s go back to Stocks L and U and assume you put half your money in each. The portfolio weights are
.50 and .50. What is the pattern of returns on this portfolio? The expected return?
To answer these questions, suppose the economy actually enters a recession. In this case, half your money
(the half in L) loses 20%. The other half (the half in U) gains 30%. Your portfolio return, RP, in a
recession is thus:
R P = (.50 × −20%) + (.50 × 30%) = 5%

Table 13.5 summarizes the remaining calculations. Notice that when a boom occurs, your portfolio
would return 40%:
R P = (.50 × 70%) + (.50 × 10%) = 40%

TABLE 13.5

Expected return on an equally weighted portfolio of Stock L and Stock U

(1) State of (2) P robability of (3) P ortfolio Return if (4) P roduct

Economy State of Economy State Occurs (2) × (3)
Recession .50 (.50 × −20%) + (.50 × 30%) = 5% 2.5%
Boom .50 (.50 × 70%) + (.50 × 10%) = 40% 20.0
E(RP) = 22.5%
 As indicated in Table 13.5, the expected return on your portfolio, E(RP), is 22.5%.
We can save ourselves some work by calculating the expected return more directly. Given these portfolio
weights, we could have reasoned that we expect half of our money to earn 25% (the half in L) and half of our
money to earn 20% (the half in U). Our portfolio expected return is:
E(R P ) = (.50 × E(R L )) + (.50 × E(R U ))

= (.50 × 25%) + (.50 × 20%) = 22.5%

This is the same portfolio expected return we had before.

This method of calculating the expected return on a portfolio works no matter how many assets are in the
portfolio. Suppose we had n assets in our portfolio, where n is any number. If we let xi stand for the
percentage of our money in asset i, the expected return is:

E(R P ) = x 1 × E(R 1 ) + x 2 × E(R 2 ) + … + x n × E(R n ) [13.4]

Page 474

This says that the expected return on a portfolio is a weighted combination of the expected returns on the
assets in that portfolio. This seems somewhat obvious, but, as we examine next, the obvious approach is
not always the right one.

XAMPLE 13.3
Portfolio Expected Returns
Suppose we have the following projections on three stocks:

Returns
State of Economy P robability of State Stock A Stock B Stock C
Boom .40 10% 15% 20%
Bust .60 8% 4% 0%

What would be the expected return on a portfolio with equal amounts invested in each of
the three stocks? What would the expected return be if half the portfolio were in A, with the
remainder equally divided between B and C?
From our earlier discussions, the expected returns on the individual stocks are (check these
for practice):
E (R A ) = 8.8%

E (R B ) = 8.4%

E (R C ) = 8.0%

If a portfolio has equal investments in each asset, the portfolio weights are all the same.
Such a portfolio is said to be equally weighted. Since there are three stocks, the weights
are all equal to 1/3. Therefore portfolio expected return is:
1 1 1
E (R P ) = ( × 8.8%) + ( × 8.4%) + ( × 8.0%)
3 3 3

= 8.4%
 In the second case, check that the portfolio expected return is 8.5%.

Portfolio Variance
From our previous discussion, the expected return on a portfolio that contains equal investment in Stocks
U and L is 22.5%. What is the standard deviation of return on this portfolio? Simple intuition might suggest
that half the money has a standard deviation of 45% and the other half has a standard deviation of 10%, so the
portfolio’s standard deviation might be calculated as:
σ P = (. 50 × 45%) + (.50 × 10%) = 27.5%

Unfortunately, this approach is incorrect.

Page 475

Let’s see what the standard deviation really is. Table 13.6 summarizes the relevant calculations. As we
see, the portfolio’s variance is about .031, and its standard deviation is less than we thought—it’s only
17.5%. What is illustrated here is that the variance on a portfolio is not generally a simple weighted
combination of the variances of the assets in the portfolio.

TABLE 13.6

Variance on an equally weighted portfolio of Stock L and Stock U

(1) State of (2) P robability of (3) P ortfolio Return (4) Squared Dev iation from Expected (5) P roduct
Economy State of Economy if State Occurs Return (2) × (4)
Recession .50 5% (.05 − .225) 2 = .030625 .0153125
Boom .50 40% (.40 − .225) 2 = .030625 .0153125

2
σ P = .030625

σ P = √ .030625 = 17.5%

We can illustrate this point a little more dramatically by considering a slightly different set of portfolio
weights. Suppose we put 2/11 (about 18%) in L and the remaining 9/11 (about 82%) in U. If a recession
occurs, this portfolio would have a return of:
2 9
RP = ( × −20%) + ( × 30%) = 20.91%
11 11

If a boom occurs, this portfolio would have a return of:

2 9
Rp = ( × 70%) + ( × 10%) = 20.91%
11 11

Notice that the return is the same no matter what happens. No further calculations are needed. This
portfolio has a zero variance. Apparently, combining assets into portfolios can substantially alter the risks
faced by the investor. This is a crucial observation, and we explore its implications in the next section.

XAMPLE 13.4
 Portfolio Variance and Standard Deviation
In Example 13.3, what are the standard deviations on the two portfolios? To answer, we
first have to calculate the portfolio returns in the two states. We will work with the second
portfolio, which has 50% in Stock A and 25% in each of Stocks B and C. The relevant
calculations can be summarized as follows:

Returns
State of Economy P robability of State Stock A Stock B Stock C Portfolio
Boom .40 10% 15% 20% 13.75%
Bust .60 8% 4% 0% 5.00%

When the economy booms, the portfolio return is calculated as:

(.50 × 10%) + (.25 × 15%) + (.25 × 20%) = 13.75%

The return when the economy goes bust is calculated the same way. The expected return
on the portfolio is (.40 × 13.75%) + (.60 × 5.00%) = 8.5% with a variance of:

σ 2 2 2
= [.40 × (.1375 − .085) ] + [.60 × (.05 − .085) ]

= .0018375

The standard deviation is about 4.3%. For our equally weighted portfolio, check to see that
the standard deviation is about 5.5%.

Portfolio Standard Deviation and Diversification

How diversification reduces portfolio risk as measured by the portfolio standard deviation is worth
exploring in some detail. 3 The key concept is correlation, which provides a measure on the extent to which
the returns on two assets move together. If correlation is positive, we say that Assets A and B are positively
correlated; if it is negative, we say they are negatively correlated; and if it is zero, the two assets are
uncorrelated.
Page 476

Figure 13.1 shows these three benchmark cases for two assets, A and B. The graphs on the left side plot
the separate returns on the two securities through time. Each point on the graphs on the right side
represents the returns for both A and B over a particular time interval. The figure shows examples of
different values for the correlation coefficient, CORR (RA, RB), that range from −1.0 to 1.0.
FIGURE 13.1
Examples of different correlation coeffi cients
The graphs on the left-hand side of the figure plot the separate returns on the two securities through time. Each point on the graphs on the right-hand side

represents the returns for both A and B over a particular time period.

To show how the graphs are constructed, we need to look at points 1 and 2 (on the upper left graph) and
relate them to point 3 (on the upper right graph). Point 1 is a return on Company B and point 2 is a return
 on Company A. They both occur over the same time period; say, for example, the month of June. Both
returns are above average. Point 3 represents the returns on both stocks in June, with the x-axis
representing the return on A and the y-axis representing the return on B. Other dots in the upper right graph
represent the returns on both stocks in other months.
Page 477

Because the returns on Security B have bigger swings than the returns on Security A, the slope of the line in
the upper right graph is greater than one. Perfect positive correlation does not imply that the slope is
necessarily one. Rather, it implies that all points lie exactly on the line. Less than perfect positive
correlation implies a positive slope, but the points do not lie exactly on the line. An example of less than
perfect positive correlation is provided on the left side of Figure 13.2. As before, each point in the
graph represents the returns on both securities in the same month. In a graph like this, the closer the points
lie to the line, the closer the correlation is to one. In other words, a high correlation between the two
returns implies that the graph has a tight fit. 4
FIGURE 13.2
Graphs of possible relationships between two stocks

Less than perfect negative correlation implies a negative slope, but the points do not lie exactly on the line,
as shown on the right side of Figure 13.2.

XAMPLE 13.5
Correlation between Stocks U and L
What is the correlation between Stocks U and L from our earlier example if we assume the
two states of the economy are equally probable? Table 13.2 shows the returns on each
stock in recession and boom states.

Stock L Stock U
 Stock L Stock U
Recession −.20 .30
Boom .70 .10

Page 478

Figure 13.3 plots the line exactly the same way as we plotted the graphs on the right
sides of Figures 13.1 and 13.2. You can see from the figure that the line has a
negative slope and all the points lie exactly on the line. (Since we only have two outcomes
for each stock, the points must plot exactly on a straight line.) You can conclude that the
correlation between Stocks U and L is equal to −1.0.
FIGURE 13.3
Correlation between Stocks U and L

Our discussion of correlation provides us with a key building block of a formula for
portfolio standard variance and its square root, portfolio standard deviation.

σ
2
p
= x
2
L
σ
2
L
+ x
2
U
σ
2
U
+ 2x L x U CORR L,U σ L σ U
[13.5]
σp = √σ 2
p

Recall that xL and xU are the portfolio weights for Stocks U and L, respectively. CORRL,U is
the correlation of the two stocks. We can use the formula to check our previous calculation
 of portfolio standard deviation for a portfolio invested equally in each stock.
2 2 2 2
= (.50 × .45 ) + (.50 × .10 ) + (2 ×.50 × .50 × − 1.0 × .45 × .10)

= .030625

= √ .030625 = 17.5%

This is the same result we got in Table 13.6.

The Efficient Set

Suppose U and L actually have a correlation of about +0.70. The opportunity set is graphed in
Figure 13.4. In Figure 13.5, we have marked the minimum variance portfolio, MV. No risk-averse
investor would hold any portfolio with expected return below MV. For example, no such investor would
invest 100% in Stock U because such a portfolio has lower expected return and higher standard deviation
than the minimum variance portfolio. We say that portfolios such as U are dominated by the minimum
variance portfolio. (Since standard deviation is the square root of variance, the minimum variance portfolios
also have minimum standard deviations as shown in Figures 13.4 and 13.5.) Though the entire curve
from U to L is called the feasible set, investors consider only the curve from MV to L since points along
this curve provide the highest possible returns for any given risk level. This part of the curve is called the
efficient set.
Page 479

FIGURE 13.4
Opportunity sets composed of holdings in Stock L and Stock U
 FIGURE 13.5
Effi cient frontier

Page 480

XAMPLE 13.6
The Zero-Variance Portfolio
Can you find a portfolio of Stocks U and L with zero variance? Earlier, we showed that
investing 2/11 (about 18%) of the portfolio in L and the other 9/11 (about 82%) in U gives the
same expected portfolio return in either recession or boom. As a result, the portfolio
variance and standard deviation should both be zero. We can check this with our formula
for portfolio variance.
2 2
2 9 2 9
2 2 2
σ p
= ( ) × (.45) + ( ) × (.10) + 2 × ( ) × ( ) × (−1.0) × (.45) × (.10)
11 11 11 11

= .006694 + .006694 − .013388 = 0

You can see that the portfolio variance (and standard deviation) are zero because the
weights were chosen to make the negative third term exactly offset the first two positive
terms. This third term is called the covariance term because the product of the correlation
times the two security standard deviations is the covariance of L and U.*
COV L,U = CORR L,U × σ L × σ U

To explore how the portfolio standard deviation depends on correlation, Table 13.7
recalculates the portfolio standard deviation, changing the correlation between U and L, yet
keeping the portfolio weights and all the other input data unchanged. When the correlation
is perfectly negative (CORR = −1.0), the portfolio standard deviation is 0 as we just
U,
L

calculated. If the two stocks were uncorrelated (CORR = 0), the portfolio standard
L,
U

deviation becomes 11.5708%. With less than perfect positive correlation

(CORR L, = +0.7), the portfolio standard deviation is 15.0865%. Lastly, with perfect
U

positive correlation (CORR = +1.0), the portfolio standard deviation becomes

L,
U

16.3636%.

TABLE 13.7

Portfolio standard deviation and correlation

Stock L xL = 2/11 σL = 45% E(RL) = 25%

Stock U xU = 9/11 σU = 10% E(RU) = 20%
E(RP) = (2/11) × 25% + 9/11 × 20% = 20.91%

CORRL,U P ortfolio Standard Dev iation of P ortfolio σP

1. −1.0 0.0000%
2. 0.0 11.5708%
3. +0.7 15.0865%
4. +1.0 16.3636%

When the returns on the two assets are perfectly correlated, the portfolio standard
deviation is simply the weighted average of the individual standard deviations. In this
special case:
2 9
( × 45%) + ( × 10%) = 16. 3636%
11 11

Page 481

With perfect correlation, all possible portfolios lie on a straight line between U and L in
expected return/standard deviation space as shown in Figure 13.4. In this polar case,
there is no benefit from diversification. But, as soon as correlation is less than perfectly
positive (CORRL,U < +1.0), diversification reduces risk.
As long as CORR is less than +1.0, the standard deviation of a portfolio of two securities is
less than the weighted average of the standard deviations of the individual securities.
Figure 13.4 shows this important result by graphing all possible portfolios of U and L
for the four cases for CORRL,U given in Table 13.7. The portfolios marked 1, 2, 3, and 4
in Figure 13.4 all have an expected return of 20.91%, as calculated in Table 13.7.
 Their standard deviations also come from Table 13.7. The other points on the respective
lines or curves are derived by varying the portfolio weights for each value of CORRL,U.
Each line or curve represents all the possible portfolios of U and L for a given correlation.
Each is called an opportunity set or feasible set. The lowest opportunity set representing
CORRL,U = 1.0 always has the largest standard deviation for any return level. Once again,
this shows how diversification reduces risk as long as correlation is less than perfectly
positive.

Correlations in the Financial Crisis of 2007–2009

During the financial crisis, investors sought safety and liquidity, causing equity markets to fall in all
countries. The resulting increase in the correlation of returns across countries led some investors to doubt
the benefits of international diversification. While higher correlations reduce its advantages, doubts about
diversification were overstated for two reasons. First, although correlations undoubtedly increased during
the crisis, they later returned to more normal levels. Second, even with relatively high positive correlation
between assets, diversification still reduces risk as long as the correlation coefficient is less than 1.0
(perfect positive correlation).

Concept Questions
. What is a portfolio weight?
. How do we calculate the expected return on a portfolio?
. Is there a simple relationship between the standard deviation on a portfolio and the
standard deviation of the assets in the portfolio?
. For diversification purposes, why must correlation between individual assets in a portfolio
be below 1.0?
 LO3 13.3 | Announcements, Surprises, and
Expected Returns
Now that we know how to construct portfolios and evaluate their returns, we begin to describe more
carefully the risks and returns associated with individual securities. So far, we have measured volatility by
looking at the differences between the actual returns on an asset or portfolio, R, and the expected return,
E(R). We now look at why those deviations exist.
Page 482

Expected and Unexpected Returns

To begin, for concreteness, we consider the return on the stock of TransCanada Industries. What will
determine this stock’s return in, say, the coming year? The return on any stock traded in a financial market is
composed of two parts. First, the normal or expected return from the stock is the part of the return that
shareholders in the market predict or expect. This return depends on the information shareholders have that
bears on the stock, and it is based on the market’s understanding today of the important factors that
influence the stock in the coming year.
The second part of the return on the stock is the uncertain or risky part. This is the portion that comes
from unexpected information that is revealed within the year. A list of all possible sources of such
information is endless, but here are a few examples:
News about TransCanada’s research
Government figures released on gross national product (GNP)
The imminent bankruptcy of an important competitor
News that TransCanada’s sales figures are higher than expected
A sudden, unexpected drop in interest rates

Based on this discussion, one way to write the return on TransCanada’s stock in the coming year would be:

Total return = Expected return + Unexpected return [13.6]

R = E(R) + U

where R stands for the actual total return in the year, E(R) stands for the expected part of the return, and U
stands for the unexpected part of the return. What this says is that the actual return, R, differs from the
expected return, E(R), because of surprises that occur during the day.

Announcements and News

We need to be careful when we talk about the effect of news items on the return. For example, suppose that
TransCanada Industries’ business is such that the company prospers when GNP grows at a relatively high
rate and suffers when GNP is relatively stagnant. In deciding what return to expect this year from owning
stock in TransCanada, shareholders either implicitly or explicitly must think about what the GNP is likely
to be for the year.
When the government actually announces GNP figures for the year, what will happen to the value of
TransCanada Industries stock? Obviously, the answer depends on what figure is released. More to the point,
however, the impact depends on how much of that figure is new information.
At the beginning of the year, market participants have some idea or forecast of what the yearly GNP will be.
To the extent that shareholders had predicted the GNP, that prediction is already factored into the expected
part of the return on the stock, E(R). On the other hand, if the announced GNP is a surprise, the effect is
part of U, the unanticipated portion of the return. As an example, suppose shareholders in the market had
forecast that the GNP increase this year would be 0.5%. If the actual announcement this year is exactly
0.5%, the same as the forecast, the shareholders didn’t really learn anything, and the announcement isn’t
news. There would be no impact on the stock price as a result. This is like receiving confirmation of
something that you suspected all along; it doesn’t reveal anything new.
Page 483

To give a more concrete example, on April 30, 2015, Tesla Motors Inc. announced its expansion into the
home battery business. This announcement seems to be great news for Tesla, but the share price dropped
by 0.99% the day after the announcement. Why? Because the company’s entry into the home battery market
was anticipated and this gave market participants time to re-price Tesla shares. Tesla’s share price had risen
by more than 20% in the month prior to the announcement.
A common way of saying that an announcement isn’t news is to say that the market has already “discounted”
the announcement. The use of the word discount here is different from the use of the term in computing
present values, but the spirit is the same. When we discount a dollar in the future, we say it is worth less to
us because of the time value of money. When we discount an announcement or a news item, we mean it has
less of an impact on the market because the market already knew much of it.
For example, going back to TransCanada Industries, suppose that the government announced that the
actual GNP increase during the year was 1.5%. Now shareholders have learned something; namely, that the
increase is one percentage point higher than had been forecast. This difference between the actual result and
the forecast, 1 percentage point in this example, is sometimes called the innovation or the surprise.
An announcement, then, can be broken into two parts—the anticipated, or expected, part and the surprise,
or innovation:

Announcement = Expected part + Surprise [13.7]

The expected part of any announcement is the part of the information that the market uses to form the
expectation, E(R), of the return on the stock. The surprise is the news that influences the unanticipated
return on the stock, U.
Our discussion of market efficiency in the previous chapter bears on this discussion. We are assuming that
relevant information that is known today is already reflected in the expected return. This is identical to
saying that the current price reflects relevant publicly available information. We are implicitly assuming that
markets are at least reasonably efficient in the semistrong form sense.
Going forward, when we speak of news, we mean the surprise part of an announcement and not the portion
that the market has expected and already discounted.
Concept Questions
. What are the two basic parts of a return?
. Under what conditions does an announcement have no effect on common stock prices?
 LO3 13.4 | Risk: Systematic and Unsystematic
The unanticipated part of the return, that portion resulting from surprises, is the true risk of any investment.
After all, if we always receive exactly what we expect, the investment is perfectly predictable and, by
definition, risk-free. In other words, the risk of owning an asset comes from surprises—unanticipated events.
Page 484

There are important differences, though, among various sources of risk. Look back at our previous list of
news stories. Some of these stories are directed specifically at TransCanada Industries, and some are more
general. Which of the news items are of specific importance to TransCanada Industries?
Announcements about interest rates or GNP are clearly important for nearly all companies, whereas the
news about TransCanada Industries’ president, its research, or its sales are of specific interest to
TransCanada Industries. We distinguish between these two types of events, however, because, as we shall
see, they have very different implications.

Systematic and Unsystematic Risk

The first surprise, the one that affects a large number of assets, we label systematic risk. A systematic risk is
one that influences a large number of assets, each to a greater or lesser extent. Because systematic risks are
market-wide effects, they are sometimes called market risks.
The second type of surprise we call unsystematic risk. An unsystematic risk is one that affects a single asset
or a small group of assets. Because these risks are unique to individual companies or assets, they are
sometimes called unique or asset-specific risks. We use these terms interchangeably.
As we have seen, uncertainties about general economic conditions, such as GNP, interest rates, or inflation,
are examples of systematic risks. These conditions affect nearly all companies to some degree. An
unanticipated increase or surprise in inflation, for example, affects wages and the costs of the supplies that
companies buy, the value of the assets that companies own, and the prices at which companies sell their
products. Forces such as these, to which all companies are susceptible, are the essence of systematic risk.
In contrast, the announcement of an oil strike by a company primarily affects that company and, perhaps, a
few others (such as primary competitors and suppliers). It is unlikely to have much of an effect on the
world oil market, however, or on the affairs of companies not in the oil business.

Systematic and Unsystematic Components of Return

The distinction between a systematic risk and an unsystematic risk is never really as exact as we make it out
to be. Even the most narrow and peculiar bits of news about a company ripple through the economy. This
is true because every enterprise, no matter how tiny, is a part of the economy. It’s like the tale of a kingdom
that was lost because one horse lost a shoe. This is mostly hairsplitting, however. Some risks are clearly
much more general than others. We’ll see some evidence on this point in just a moment.
The distinction between the types of risk allows us to break down the surprise portion, U, of the return on
TransCanada Industries stock into two parts. As before, we break the actual return down into its expected
and surprise components:
R = E (R) + U

We now recognize that the total surprise for TransCanada Industries, U, has a systematic and an
unsystematic component, so:

R = E (R) + Syst emat ic port ion + Unsyst emat ic port ion [13.8]

Page 485

Because it is traditional, we use the Greek letter epsilon, ε, to stand for the unsystematic portion. Since
systematic risks are often called market risks, we use the letter m to stand for the systematic part of the
surprise. With these symbols, we can rewrite the total return:
R = E (R) + U
= E (R) + m + ε

The important thing about the way we have broken down the total surprise, U, is that the unsystematic
portion, ε, is more or less unique to TransCanada Industries. For this reason, it is unrelated to the
unsystematic portion of return on most other assets. To see why this is important, we need to return to the
subject of portfolio risk.

Concept Questions
. What are the two basic types of risk?
. What is the distinction between the two types of risk?
 LO2 13.5 | Diversification and Portfolio Risk
We’ve seen earlier that portfolio risks can, in principle, be quite different from the risks of the assets that
make up the portfolio. We now look more closely at the riskiness of an individual asset versus the risk of a
portfolio of many different assets. We once again examine some market history to get an idea of what
happens with actual investments in capital markets.

The Effect of Diversification: Another Lesson from

Market History
In our previous chapter, we saw that the standard deviation of the annual return on a portfolio of several
hundred large common stocks has historically been about 17% per year for both the Toronto Stock
Exchange and the New York Stock Exchange (see Table 12.4, for example). Does this mean the
standard deviation of the annual return on a typical stock is about 17%? As you might suspect by now, the
answer is no. This is an extremely important observation.

To examine the relationship between portfolio size and portfolio risk, Table 13.8 illustrates typical
average annual standard deviations for equally weighted portfolios that contain different numbers of
randomly selected NYSE securities. 5

TABLE 13.8

Standard deviations of annual portfolio returns

(1) Number of Stocks (2) Av erage Standard Dev iation of (3) Ratio of P ortfolio Standard Dev iation to Standard
in P ortfolio Annual P ortfolio Returns Dev iation of a Single Stock
1 49.24% 1.00
2 37.36 0.76
4 29.69 0.60
6 26.64 0.54
8 24.98 0.51
10 23.93 0.49
20 21.68 0.44
30 20.87 0.42
40 20.46 0.42
50 20.20 0.41
100 19.69 0.40
200 19.42 0.39
300 19.34 0.39
400 19.29 0.39
500 19.27 0.39
 (1) Number of Stocks (2) Av erage Standard Dev iation of (3) Ratio of P ortfolio Standard Dev iation to Standard
in P ortfolio Annual P ortfolio Returns Dev iation of a Single Stock
1,000 19.21 0.39

Page 486

In Column 2 of Table 13.8, we see that the standard deviation for a “portfolio” of one security is about
49%. What this means is that, if you randomly selected a single NYSE stock and put all your money into it,
your standard deviation of return would typically have been a substantial 49% per year. If you were to
randomly select two stocks and invest half your money in each, your standard deviation would have been
about 37% on average, and so on.

The important thing to notice in Table 13.8 is that the standard deviation declines as the number of
securities increases. By the time we have 30 randomly chosen stocks, the portfolio’s standard deviation has
declined by about 60%, from 49 to about 20%. With 500 securities, the standard deviation is 19.27%,
similar to the 21% we saw in our previous chapter for the large common stock portfolio. The small
difference exists because the portfolio securities and time periods examined are not identical.

The Principle of Diversification

Figure 13.6 illustrates the point we’ve been discussing. What we have plotted is the standard deviation of
return versus the number of stocks in the portfolio. Notice in Figure 13.6 that the benefit in risk
reduction from adding securities drops as we add more and more. By the time we have ten securities, the
portfolio standard deviation has dropped from 49.2 to 23.9%, most of the effect is already realized, and by
the time we get to 30 or so, there is very little remaining benefit. The data in Table 13.8 and
Figure 13.6 is from the NYSE, but a Canadian study documented the same effect. However, the
Canadian researchers found that Canadian investors need to hold a larger number of stocks to achieve
diversification. This is likely due to Canadian stocks being more concentrated in a few industries than in the
U.S. 6
Page 487

FIGURE 13.6
P ortfolio div ersifi cation
 Figure 13.6 illustrates two key points. First, the principle of diversification (discussed earlier) tells us
that spreading an investment across many assets eliminates some of the risk. The shaded area in
Figure 13.6, labelled diversifiable risk, is the part that can be eliminated by diversification.
The second point is equally important. A minimum level of risk cannot be eliminated simply by
diversifying. This minimum level is labelled nondiversifiable risk in Figure 13.6. Taken together, these
two points are another important lesson from capital market history—diversification reduces risk, but only
up to a point. Put another way, some risk is diversifiable and some not.
To give a recent example of the impact of diversification, the S&P TSX Composite Index, an index of large
Canadian stocks, plunged 37% from peak to trough due to COVID-19 pandemic impacts, then bounced
back to post a modest 2% gain in 2020. The Information Technology sector within that index was up by
80% (led by Shopify) while the Energy sector returned –31%. Again, the lesson is clear—diversification
reduces exposure to extreme outcomes, both good and bad.
Diversification and Unsystematic Risk
From our discussion of portfolio risk, we know that some of the risk associated with individual assets can
be diversified away and some cannot. We are left with an obvious question: why is this so? It turns out that
the answer hinges on the distinction we made earlier between systematic and unsystematic risk.
Page 488

By definition, an unsystematic risk is one that is particular to a single asset or, at most, a small group. For
example, if the asset under consideration is stock in a single company, the discovery of positive NPV
projects such as successful new products and innovative cost savings tend to increase the value of the
stock. Unanticipated lawsuits, industrial accidents, strikes, and similar events tend to decrease future cash
flows, thereby reducing share values.
Here is the important observation. If we held only single stock, the value of our investment would fluctuate
because of company-specific events. If we held a large portfolio, on the other hand, some of the stocks in
the portfolio would go up in value because of positive company-specific events and some would go down
in value because of negative events. The net effect on the overall value of the portfolio is relatively small,
however, as these effects tend to cancel each other out.
Now we see why some of the variability associated with individual assets is eliminated by diversification. By
combining assets into portfolios, the unique or unsystematic events—both positive and negative—tend to
wash out once we have more than just a few assets.
This important point bears repeating: Unsystematic risk is essentially eliminated by diversification, so a
relatively large portfolio has almost no unsystematic risk. In fact, the terms diversifiable risk and unsystematic
risk are often used interchangeably.

Diversification and Systematic Risk

We’ve seen that unsystematic risk can be eliminated by diversifying. What about systematic risk? Can it also
be eliminated by diversification? The answer is no because, by definition, a systematic risk affects almost all
assets to some degree. As a result, no matter how many assets we put into a portfolio, the systematic risk
doesn’t go away. The terms systematic risk and nondiversifiable risk are also used interchangeably.
Because we have introduced so many different terms, it is useful to summarize our discussion before
moving on. What we have seen is that the total risk of an investment, as measured by the standard deviation
of its return, can be written as:

Total risk = Systematic risk + Unsystematic risk [13.9]

Systematic risk is also called nondiversifiable risk or market risk. Unsystematic risk is also called
diversifiable risk, unique risk, or asset-specific risk. For a well-diversified portfolio, the unsystematic risk is
negligible. For such a portfolio, essentially all of the risk is systematic.

Risk and the Sensible Investor

 Having gone to all this trouble to show that unsystematic risk disappears in a well-diversified portfolio, how
do we know that investors even want such portfolios? Suppose they like risk and don’t want it to disappear?
We must admit that, theoretically at least, this is possible, but we argue that it does not describe what we
think of as the typical investor. Our typical investor is risk averse. Risk-averse behaviour can be defined in
many ways, but we prefer the following example. A fair gamble is one with zero expected return; a risk-
averse investor would prefer to avoid fair gambles.
Why do investors choose well-diversified portfolios? Our answer is that they are risk averse, and risk-averse
people avoid unnecessary risk, such as the unsystematic risk on a stock. If you do not think this is much of
an answer to why investors choose well-diversified portfolios and avoid unsystematic risk, consider whether
you would take on such a risk. For example, suppose you had worked all summer and had saved $5,000,
which you intended to use for your university expenses. Now, suppose someone came up to you and offered
to flip a coin for the money; heads, you would double your money, and tails, you would lose it all.
Page 489

Would you take such a bet? Perhaps you would, but most people would not. Leaving aside any moral
question that might surround gambling and recognizing that some people would take such a bet, it’s our
view that the average investor would not.
To induce the typical risk-averse investor to take a fair gamble, you must sweeten the pot. For example, you
might need to raise the odds of winning from 50–50 to 70–30 or higher. The risk-averse investor can be
induced to take fair gambles only if they are sweetened so that they become unfair to the investor’s
advantage.

XAMPLE 13.7
Risk of Canadian Mutual Funds
Table 13.9 shows the returns and standard deviations for two Canadian mutual funds
over the three-year period ending December 31, 2017. The table also shows comparable
statistics for the S&P/TSX Composite. As you would expect, the TSX portfolio is the most
widely diversified of the three portfolios and has the lowest unsystematic risk. For this
reason, it has the lowest portfolio standard deviation. The next lowest standard deviation is
the CIBC Canadian Equity fund, which invests in equities across different Canadian
industries.

TABLE 13.9

Average returns and standard deviations for two Canadian mutual funds and S&P/TSX Composite, 2015–2017

Fund Annual Return (% ) Standard Dev iation (% )

S&P/TSX Composite 6.59 7.35
CIBC Canadian Equity 5.24 7.40
CIBC Precious Metals 2.69 36.89

Source: The Globe and Mail, Mutual Funds, theglobeandmail.com/globe-investor/funds-and-etfs/funds.

The CIBC Precious Metals fund focuses on one sector of the economy. For example, its top
three holdings at the end of December 2017 were Newmont Mining, Franco-Nevada Corp.,
and Agnico Eagle Mines. The narrower focus of this fund makes it less diversified, with
higher standard deviations.
What does this example tell us about how good these funds were as investments? To
answer this question, we have to investigate asset pricing, our next topic.

Concept Questions
. What happens to the standard deviation of return for a portfolio if we increase the number
of securities in the portfolio?
. What is the principle of diversification?
. Why is some risk diversifiable? Why is some risk not diversifiable?
. Why can’t systematic risk be diversified away?
. Explain the concept of risk aversion.
 Page 490

LO4 13.6 | Systematic Risk and Beta

The question that we now begin to address is, what determines the size of the risk premium on a risky asset?
Put another way, why do some assets have a larger risk premium than other assets? The answer to these
questions, as we discuss next, is also based on the distinction between systematic and unsystematic risk.

The Systematic Risk Principle

We’ve seen that the total risk associated with an asset can be decomposed into two components: systematic
and unsystematic risk. We have also seen that unsystematic risk can be essentially eliminated by
diversification. The systematic risk present in an asset, on the other hand, cannot be eliminated by
diversification.
Based on our study of capital market history, we know that there is a reward, on average, for bearing risk.
However, we now need to be more precise about what we mean by risk. The systematic risk principle states
that the reward for bearing risk depends only on the systematic risk of an investment. The underlying
rationale for this principle is straightforward—because unsystematic risk can be eliminated at virtually no
cost (by diversifying), there is no reward for bearing it. Put another way, the market does not reward risks
that are borne unnecessarily.
The systematic risk principle has a remarkable and very important implication; the expected return on an
asset depends only on that asset’s systematic risk. There is an obvious corollary to this principle. No matter
how much total risk an asset has, only the systematic portion is relevant in determining the expected return
(and the risk premium) on that asset.

Measuring Systematic Risk

Since systematic risk is the crucial determinant of an asset’s expected return, we need some way of
measuring the level of systematic risk for different investments. The specific measure that we use is called
the beta coefficient, for which we will use the Greek symbol β. A beta coefficient, or beta for short, tells us
how much systematic risk a particular asset has relative to an average asset representing the market
portfolio. By definition, an average asset has a beta of 1.0 relative to itself. An asset with a beta of .50 has
half as much systematic risk as an average asset; an asset with a beta of 2.0 has twice as much. These
different levels of beta are illustrated in Figure 13.7. You can see that high beta assets display greater
volatility over time.
FIGURE 13.7
Volatility: High and low betas
 Page 491

Table 13.10 contains the estimated beta coefficients for the stocks of some well-known companies
ranging from 0.36 to 2.77.

TABLE 13.10

Beta coefficients for selected companies

Companies Beta Coeffi cient

Bank of Nova Scotia 0.88
IGM Financial 1.70
Enbridge Inc. 0.94
Manulife Financial Corp 1.25
Rogers Communications 0.33
Teck Resources Ltd. 1.44

Source : Ya hoo Fina nce https://fina nce .ya hoo.com/.

XAMPLE 13.8
Total Risk versus Beta
Consider the following information on two securities. Which has greater total risk? Which
has greater systematic risk? Greater unsystematic risk? Which asset has a higher risk
premium?

Standard Dev iation Beta

 Standard Dev iation Beta
Security A 40% .50
Security B 20 1.50

From our discussion in this section, Security A has greater total risk, but it has substantially
less systematic risk. Since total risk is the sum of systematic and unsystematic risk, Security
A must have greater unsystematic risk. Finally, from the systematic risk principle, Security B
has a higher risk premium and a greater expected return, despite the fact that it has less
total risk.

Remember that the expected return, and thus the risk premium, on an asset depends only on its systematic
risk. Because assets with larger betas have greater systematic risks, they have greater expected returns. Thus,
in Table 13.10, an investor who buys stock in Bank of Nova Scotia, with a beta of 0.88, should expect
to earn less, on average, than an investor who buys stock in IGM Financial, with a beta of 1.70.

Portfolio Betas
Earlier, we saw that the riskiness of a portfolio does not have any simple relationship to the risks of the
assets in the portfolio. A portfolio beta, however, can be calculated just like a portfolio expected return.
For example, looking at Table 13.10, suppose you put half of your money in Bank of Nova Scotia and
half in IGM Financial. What would the beta of this combination be? Since Bank of Nova Scotia (BNS) has
a beta of 0.88 and Teck Resources Ltd. (TRL) has a beta of 1.70, the portfolio’s beta, βP, would be:
βP = (.50 ×β BN S ) + (.50 × β T RL )

= (.50 × 0. 88) + (.50 × 1. 70)

= 1. 29

Page 492

In general, if we had a large number of assets in a portfolio, we would multiply each asset’s beta by its
portfolio weight and then add the results to get the portfolio’s beta.

XAMPLE 13.9
Portfolio Betas
Suppose we had the following investments:

Security Amount Inv ested Expected Return Beta

Stock A $1,000 8% .80
Stock B 2,000 12 .95
Stock C 3,000 15 1.10
Stock D 4,000 18 1.40

What is the expected return on this portfolio? What is the beta of this portfolio? Does this
portfolio have more or less systematic risk than an average asset?
To answer, we first have to calculate the portfolio weights. Notice that the total amount
invested is $10,000. Of this, $1,000/$10,000 = 10% is invested in Stock A. Similarly, 20% is
invested in Stock B, 30% is invested in Stock C, and 40% is invested in Stock D. The
expected return, E(RP), is:
E(R P ) = .10 × E(R A ) + .20 × E(R B ) + .30 × E(R C ) + .40 × E(R D )

= (.10 × 8%) + (.20 × 12%) + (.30 × 15%) + (.40 × 18%) = 14.9%

Similarly, the portfolio beta, βP, is:

βP = .10 × β A + .20 × β B + .30 × β C + .40 × β D

= (.10 × .80) + (. 20 × .95) + (.30 × 1.10) + (.40 × 1.40) = 1.16

This portfolio thus has an expected return of 14.9% and a beta of 1.16. Since the beta is
larger than 1.0, this portfolio has greater systematic risk than an average asset.

Concept Questions
. What is the systematic risk principle?
. What does a beta coefficient measure?
. How do you calculate a portfolio beta?
. Does the expected return on a risky asset depend on that asset’s total risk? Explain.
 Page 493

LO4 13.7 | The Security Market Line

We’re now in a position to see how risk is rewarded in the marketplace. To begin, suppose Asset A has an
expected return of E (R ) = 20% and a beta of β = 1.6. Furthermore, the risk-free rate is R = 8% ,
A A f

the return on three-month Treasury bills. We choose this measure because it matches the investor’s horizon
for measuring performance. Notice that a risk-free asset, by definition, has no systematic risk (or
unsystematic risk), so a risk-free asset has a beta of 0.

Beta and the Risk Premium

Consider a portfolio made up of Asset A and a risk-free asset. We can calculate some different possible
portfolio expected returns and betas by varying the percentages invested in these two assets. For example, if
25% of the portfolio is invested in Asset A, the expected return is:
E(R P ) = .25 × E(R A ) + (1 − .25) × R f

= (.25 × 20%) + (.75 × 8%)

= 11.0%

Similarly, the beta on the portfolio, βP, would be:

βP = .25 × β A + (1 − .25) × 0

= .25 × 1.6

= .40

Notice that, since the weights have to add up to 1, the percentage invested in the risk-free asset is equal to 1
minus the percentage invested in Asset A.
One thing that you might wonder about is whether it is possible for the percentage invested in Asset A to
exceed 100%. The answer is yes. This can happen if the investor borrows at the risk-free rate. For example,
suppose an investor has $100 and borrows an additional $50 at 8%, the risk-free rate. The total investment
in Asset A would be $150, or 150% of the investor’s wealth. The expected return in this case would be:
E(R P ) = 1.50 × E(R A ) + (1 − 1.50) × R f

= (1.50 × 20%) − (.50 × 8%)

= 26%

The beta on the portfolio would be:

βP = 1.50 × β A + (1 − 1.50) × 0

= 1.50 × 1.6

= 2.4

We can calculate some other possibilities as follows:

P ercentage of P ortfolio in Asset A P ortfolio Expected Return P ortfolio Beta

0% 8% 0.0
 P ercentage of P ortfolio in Asset A P ortfolio Expected Return P ortfolio Beta
25 11 0.4
50 14 0.8
75 17 1.2
100 20 1.6
125 23 2.0
150 26 2.4

Page 494

In Figure 13.8A, the portfolio expected returns are plotted against the portfolio betas. Notice that all
the combinations fall on a straight line.
FIGURE 13.8A
P ortfolio expected returns and betas for Asset A

THE REWARD-TO-RISK RATIO

What is the slope of the straight line in Figure 13.8A? As always, the slope of a straight line is equal to
“the rise over the run.” As we move out of the risk-free asset into Asset A, the beta increases from zero to
1.6 (a “run” of 1.6). At the same time, the expected return goes from 8 to 20%, a “rise” of 12%. The slope
of the line is thus12%/1.6 = 7.50% .
Notice that the slope of our line is just the risk premium on Asset A, E (R A ) − R f , divided by Asset A’s
beta, β A:
E(R A ) − R f
Slope =
βA

20% − 8%
= = 7.50%
1.6
What this tells us is that Asset A offers a reward-to-risk ratio of 7.50%. 7 In other words, Asset A has a risk
premium of 7.50% per unit of systematic risk.

THE BASIC ARGUMENT

Now suppose we consider a second asset, Asset B. This asset has a beta of 1.2 and an expected return of
16%. Which investment is better, Asset A or Asset B? You might think that, once again, we really cannot
say. Some investors might prefer A; some investors might prefer B. Actually, however, we can say A is
better because, as we demonstrate, B offers inadequate compensation for its level of systematic risk, at least
relative to A.
To begin, we calculate different combinations of expected returns and betas for portfolios of Asset B and a
risk-free asset just as we did for Asset A. For example, if we put 25% in Asset B and the remaining 75% in
the risk-free asset, the portfolio’s expected return would be:
E(R P ) = .25 × E(R B ) + (1 − .25) × R f

= (.25 × 16%) + (.75 × 8%)

= 10%

Page 495

Similarly, the beta on the portfolio, βP, would be:

βP = .25 × β B + (1 − .25) × 0

= .25 × 1.2

= .30

Some other possibilities are as follows:

P ercentage of P ortfolio in Asset B P ortfolio Expected Return P ortfolio Beta

0% 8% 0.0
25 10 0.3
50 12 0.6
75 14 0.9
100 16 1.2
125 18 1.5
150 20 1.8

When we plot these combinations of portfolio expected returns and portfolios betas in Figure 13.8B, we
get a straight line just as we did for Asset A.
FIGURE 13.8B
P ortfolio expected returns and betas for Asset B
The key thing to notice is that when we compare the results for Assets A and B, as in Figure 13.8C, the
line describing the combinations of expected returns and betas for Asset A is higher than the one for Asset
B. This tells us that for any given level of systematic risk (as measured by β), some combination of Asset A
and the risk-free asset always offers a larger return. This is why we were able to state that Asset A is a better
investment than Asset B.
Page 496
FIGURE 13.8C
P ortfolio expected returns and betas for both assets
Another way of seeing that A offers a superior return for its level of risk is to note that the slope of our line
for Asset B is:
E(R B ) − R f
Slope =
βB

16% − 8%
= = 6.67%
1.2

Thus, Asset B has a reward-to-risk ratio of 6.67%, which is less than the 7.5% offered by Asset A.

THE FUNDAMENTAL RESULT

The situation we have described for Assets A and B cannot persist in a well-organized, active market,
because investors would be attracted to Asset A and away from Asset B. As a result, Asset A’s price would
rise and Asset B’s price would fall. Since prices and returns move in opposite directions, the result is that
A’s expected return would decline and B’s would rise.
This buying and selling would continue until the two assets plotted on exactly the same line, which means
they offer the same reward for bearing risk. In other words, in an active, competitive market, we must have:
E(R A ) − R f E(R B ) − R f
=
βA βB

This is the fundamental relationship between risk and return.

Our basic argument can be extended to more than just two assets. In fact, no matter how many assets we
had, we would always reach the same conclusion:
The reward-to-risk ratio must be the same for all the assets in the market.
This result is really not so surprising. What it says, for example, is that, if one asset has twice as much
systematic risk as another asset, its risk premium is simply twice as large.
Since all the assets in the market must have the same reward-to-risk ratio, they all must plot on the same line
in market equilibrium. This argument is illustrated in Figure 13.9. As shown, Assets A and B plot
directly on the line and thus have the same reward-to-risk ratio. If an asset plotted above the line, such as C
in Figure 13.9, its price would rise, and its expected return would fall until it plotted exactly on the line.
Similarly, if an asset plotted below the line, such as D in Figure 13.9, its expected return would rise until
it too plotted directly on the line.
FIGURE 13.9
Expected returns and systematic risk
 The fundamental relationship between beta and expected return is that all assets must have the same reward-to-risk ratio [E(Ri) − Rf ]/βi. This means they would

all plot on the same straight line. Assets A and B are examples of this behaviour. Asset C’s expected return is too high; Asset D’s is too low.

Page 497

The arguments we have presented apply to active, competitive, well-functioning markets. The financial
markets, such as the TSX, NYSE, and NASDAQ, best meet these criteria. Other markets, such as real asset
markets, may or may not. For this reason, these concepts are most useful in examining financial markets.
We thus focus on such markets here. However, as we discuss in a later section, the information about risk
and return gleaned from financial markets is crucial in evaluating the investments that a corporation makes
in real assets.

XAMPLE 13.10
Determining Portfolio Weights
Suppose you are a risk-averse investor and you want to construct an investment portfolio
with 5% standard deviation. There are two assets in the market you can choose from: a
riskless Government of Canada T-bill with expected return of 3%, and a risky S&P/TSX
index fund with expected return of 10% and standard deviation of 16%. What are the
portfolio weights? What is the expected return of the portfolio?
We can use Equation 13.5 to calculate the weights of two assets:
2 2 2 2 2
(5%) = x × (0%) + x × (16%) + 2 × x T × x S × 0 × (0%) × (20%)
T S

2
0.0025 = x × 0.0256
S

2
x = 0.09766
S

xS = 31.25%

xT = 1 − x S = 68.75%
 You will allocate 68.75% of your investment in T-bills and 31.25% in the S&P/TSX index fund.
The portfolio has an expected return of 5.19% = 3% × 68.75% + 10% × 31.25%.

Page 498

XAMPLE 13.11
Beta and Stock Valuation
An asset is said to be overvalued if its expected return is too low given its risk. Suppose
you observe the following situation:

Security Beta Expected Return

Market Portfolio 1 12%
Insec Company .8 10

The risk-free rate is currently 6%. Is Insec overvalued or undervalued relative to the market
portfolio?
To answer, we compute the reward-to-risk ratio for both. For Insec, this ratio is
(10% − 6%)/0.8 = 5%. For the market, this ratio is 6%. What we conclude is that Insec

offers an insufficient expected return for its level of risk. Insec plots below the security
market line. As investors are attracted to the market portfolio and away from Insec, Insec’s
share price would drop. As a result, Insec’s expected return would rise until the security has
the same reward-to-risk ratio as the market portfolio. To see why this is true, recall that the
dividend valuation model presented in Chapter 8 treats price as the present value of
future dividends.
D1
P0 =
r − g

Projecting the dividend stream gives us D1 and g. If the required rate of return is too low,
the stock price will be too high. For example, suppose D = $2.00 and g = 7%. If the
1

expected rate of return on the stock is wrongly underestimated at 10%, the stock price
estimate is $66.67. This price is too high if the true expected rate of return is 10.8% (we will
show how the true expected rate of return is calculated in a later section). At this higher rate
of return, the stock price should fall to $52.63. In other words, Insec is overvalued, and we
would expect to see its price fall to $52.63.

Calculating Beta
The beta of a security measures the responsiveness of that security’s return to the return on the market as a
whole. To calculate beta, we draw a line relating the expected return on the security to different returns on
the market. This line, called the characteristic line of the security, has a slope equal to the stock’s beta.

XAMPLE 13.12
 Mutual Fund Performance

TABLE 13.11

Average returns and betas for selected Canadian mutual funds, S&P/TSX Composite, and Canadian three-month
Treasury bills, 15 years ending March 31, 2015

Fund Annual Return (% ) Three-Year Beta

S&P/TSX Composite 5.57 1
Three-month Treasury Bills 2.38 0
RBC Canadian Equity 5.52 0.98
BMO Dividend Fund 8.95 0.67

Source: The TMX Group Limited, TMX Money, https://web.tmxmoney.com.

Table 13.11 gives the inputs needed to compute the reward-to-risk ratios for the TSX
and two mutual funds. Starting with the TSX, the reward-to-risk ratio was:
Average return − Riskless rate

Beta
5.57 − 2.38
= 3.19%
1. 00

Page 499

You can verify that the reward-to-risk ratio for RBC Canadian Equity was 3.20% and the
ratio for BMO Dividend was 9.81%. BMO Dividend fund beat the market by earning a higher
reward-to-risk ratio than the TSX, while RBC Canadian Equity fund slightly outperformed
over this period. Keep in mind that it is unlikely for a mutual fund, such as RBC Canadian
Equity fund, to constantly outperform the index it tracked. This is because mutual funds
charge management fees. For example, RBC Canadian Equity charges a management fee
of 1.75%.
Unfortunately, in an efficient market, while past performance may guide expectations of
future returns, these expectations may not be realized in actual returns. So, we would not
expect that BMO Dividend fund would beat the market consistently over time.

Consider Figure 13.10, which displays returns for both a hypothetical company and the market as a
whole. 8 Each point represents a pair of returns over a particular month. The vertical dimension measures
the return on the stock over the month and the horizontal dimension measures that of the S&P/TSX
Composite. (The S&P/TSX Composite is considered a reasonable proxy for the general market.)
FIGURE 13.10
Graphic representation of beta
 Figure 13.10 also shows the line of best fit superimposed on these points. In practical applications, this
line is calculated from regression analysis. As one can see from the graph, the slope is 1.28. Because the
average beta is 1, this indicates the stock’s beta of 1.28 is higher than that for the average stock.
Page 500

The goal of a financial analyst is to determine the beta that a stock will have in the future, because this is
when the proceeds of an investment are received. Of course, past data must be used in regression analysis.
Thus, it is incorrect to think of 1.28 as the beta of our example company. Rather, it is our estimate of the
firm’s beta from past data.

The bottom of Figure 13.10 indicates that the company’s R2 over the time period is 0.584. What does
this mean? R2 measures how close the points in the figure are to the characteristic line. The highest value
for R2 is 1, a situation that would occur if all points lay exactly on the characteristic line. This would be the
case where the security’s return is determined only by the market’s return without the security having any
independent variation. The R2 is likely to approach one for a large portfolio of securities. For example,
many widely diversified mutual funds have R2s of 0.80 or more. The lowest possible R2 is zero, a situation
occurring when two variables are entirely unrelated to each other. Those companies whose returns are pretty
much independent of returns on the stock market would have R2s near zero.
The risk of any security can be broken down into unsystematic and systematic risk. Whereas beta measures
the amount of systematic risk, R2 measures the proportion of total risk that is systematic. Thus, a low R2

9
indicates that most of the risk of a firm is unsystematic. 9
The mechanics for calculating betas are quite simple. People in business frequently estimate beta by using
commercially available computer programs. Certain handheld calculators are also able to perform the
calculation. In addition, a large number of services sell or even give away estimates of beta for different
firms. For example, Table 13.10 presents a set of betas calculated by Yahoo Finance in February 2021.
In calculating betas, analysts make a number of assumptions consistent with Canadian research on the
capital asset pricing model. 10 First, they generally choose monthly data, as do many financial economists.
On the one hand, statistical problems frequently arise when time intervals shorter than a month are used. On
the other hand, important information is lost when longer intervals are employed. Thus, the choice of
monthly data can be viewed as a compromise.
Second, analysts typically use just under five years of data, the result of another compromise. Due to
changes in production mix, production techniques, management style, and/or financial leverage, a firm’s
nature adjusts over time. A long time period for calculating beta implies many out-of-date observations.
Conversely, a short time period leads to statistical imprecision, because few monthly observations are used.

Concept Questions
. What is the statistical procedure employed for calculating beta?
. Why do financial analysts use monthly data when calculating beta?
. What is R2?

Page 501

The Security Market Line

The line that results when we plot expected returns and beta coefficients is obviously of some importance,
so it’s time we gave it a name. This line, which we use to describe the relationship between systematic risk
and expected return in financial markets, is usually called the security market line (SML). After NPV, the
SML is arguably the most important concept in modern finance.

MARKET PORTFOLIOS
It will be very useful to know the equation of the SML. There are many different ways that we could write it,
but one way is particularly common. Suppose we were to consider a portfolio made up of all of the assets
in the market. Such a portfolio is called a market portfolio, and we write the expected return on this market
portfolio as E(RM).
Since all the assets in the market must plot on the SML, so must a market portfolio made of those assets.
To determine where it plots on the SML, we need to know the beta of the market portfolio, βM. Because this
portfolio is representative of all the assets in the market, it must have average systematic risk. In other
words, it has a beta of one. We could therefore write the slope of the SML as:
[E(R M ) − R f ] [E(R M ) − R f ]
SML slope = = = E(R M ) − R f
βM 1
The term E(RM) − Rf is often called the market risk premium since it is the risk premium on a market
portfolio.

THE CAPITAL ASSET PRICING MODEL

To finish up, if we let E(Ri) and βi stand for the expected return and beta, respectively, on any asset in the
market, we know it must plot on the SML. As a result, we know that its reward-to-risk ratio is the same as
the overall market’s:
[E(R i ) − R f ]
= E(R M ) − R f
βi

If we rearrange this, we can write the equation for the SML as:

E(R i ) = R f + [E(R M ) − R f ] × β i
[13.10]

This result is identical to the famous capital asset pricing model (CAPM). 11
The CAPM shows that the expected return for a particular asset depends on three things:
1. The pure time value of money. As measured by the risk-free rate, Rf, this is the reward for merely waiting
for your money, without taking any risk.
2. The reward for bearing systematic risk. As measured by the market risk premium [E(RM) − Rf], this
component is the reward the market offers for bearing an average amount of systematic risk in addition to
waiting.
3. The amount of systematic risk. As measured by β,i this is the amount of systematic risk present in a
particular asset, relative to an average asset.

Page 502

Recall that in Example 13.12, we used 10.8% as the true expected return of Insec. It is calculated using
the CAPM as follows:
E(R) = R f + [E(R M ) − R f ] × β = 6% + [6% × 0.8] = 10.8%

By the way, the CAPM works for portfolios of assets just as it does for individual assets. In an earlier
section, we saw how to calculate a portfolio’s β. To find the expected return on a portfolio, we simply use
this β in the CAPM equation.

Figure 13.11 summarizes our discussion of the SML and the CAPM. As before, we plot the expected
return against beta. Now we recognize that, based on the CAPM, the slope of the SML is equal to the
market risk premium [E(RM) − Rf]. This concludes our presentation of concepts related to the risk-return
trade-off. For future reference, Table 13.12 summarizes the various concepts in the order we discussed
them in.
Page 503
FIGURE 13.11
The security market line (SML)
The slope of the security market line is equal to the market risk premium; that is, the reward for bearing an average amount of systematic risk.

The equation describing the SML can be written:

E (R ) = R + β × [E (R ) − R ]
i f i M f

which is the capital asset pricing model (CAPM).

TABLE 13.12

Summary of risk and return

Total Risk
The total risk of an investment is measured by the variance or, more commonly, the standard deviation of its return.
Total Return
The total re turn on an investment has two components: the expected return and the unexpected return. The unexpected return comes
about because of unanticipated events. The risk from investing stems from the possibility of unanticipated events.
Systematic and Unsystematic Risks
Systematic risks (also called market risks) are unanticipated events that affect almost all assets to some degree because they are
economy wide. Unsystematic risks are unanticipated events that affect single assets or small groups of assets. Unsystematic risks are
also called unique or asse t-spe cific risks.
The Effect of Div ersifi cation
Some, but not all, of the risk associated with a risky investment can be eliminated by diversification. The reason is that unsystematic
risks, which are unique to individual assets, tend to wash out in a large portfolio; systematic risks, which affect all of the assets in a
portfolio to some extent, do not.
The Systematic Risk P rinciple and Beta
Because unsystematic risk can be freely eliminated by diversification, the syste matic risk principle states that the reward for bearing risk
depends only on the level of systematic risk. The level of systematic risk in a particular asset, relative to average, is given by the beta of
that asset.
The Reward-to-Risk Ratio and the Security Market Line
 Total Risk
The reward-to-risk ratio for asset i is the ratio of its risk premium E(Ri) − Rf to its beta, β:i
E(R i ) − R f

βi

In a well-functioning market, this ratio is the same for every asset. As a result, when asset expected returns are plotted against asset
betas, all assets plot on the same straight line, called the se curity marke t line (SML).
The Capital Asset P ricing Model
From the SML, the expected return on asset i can be written:
E(R i ) = R f + [E(R M ) − R f ] × β i

This is the capital asse t pricing mode l (CAPM). The expected return on a risky asset thus has three components: The first is the pure
time value of money (Rf), the second is the market risk premium [E(RM) − Rf], and the third is the beta for that asset, β.i

IN THEIR OWN WORDS …

Pat Chiefalo on Global Investing through Exchange-Traded Funds

Home is where the heart is, they say—and it’s also where Canadians like to keep their
investments. The so-called “home bias” among investors is strong: the typical Canadian
holds about two-thirds of assets in domestic stocks and bonds. Now, there is nothing
unusual about this; investors in most countries tend to prefer the comfort of home. But there
is good reason for Canadian investors in particular to look beyond their own borders—and
to consider using exchange-traded funds as a cost-efficient, convenient way to do it.
Let’s look at the reasons for going global. One of them is the relatively small size of the
Canadian stock market, which accounts for only about 4% of global equities. When you put
all or most of your eggs in that small basket, you’re taking on more risk than if you were
more broadly diversified. There is also a lack of diversification within the Canadian market
itself. It’s highly concentrated in just three sectors: energy, financials, and materials. With
health care, industrials, technology and consumer sectors under-represented here at home,
strong sector diversification can be hard to achieve—unless you look abroad.
Another factor is economic risk. We have seen the impact of sharply lower oil prices on the
Canadian stock market and on the broad economy, which contracted in February [2015].
With oil supply remaining stronger than demand, the oil shock does not look poised to end
any time soon. The Canadian consumer, meanwhile, is saddled with record levels of
household debt compared with disposable income, and may be overleveraged. The
housing market has been strong, but may be overheated, and there are already signs of
slowing in certain areas, particularly Alberta. Put it all together, and economic growth looks
like it could well remain low and slow for some time.
The third factor for investors to consider about going global is the opportunity to manage—
and benefit—from potential currency fluctuations with ETFs. Investors who are planning to
hold an international fund for a short period and are looking to benefit from a stronger
dollar relative to the foreign currency, or who simply want to manage risk, should consider
their hedging options. For example, Canadians looking to benefit from a strong U.S. dollar
may want to add an unhedged U.S. fund, or when looking to Europe, may want to hedge
their exposure to seek potential gains from the drop in currency. Depending on your view,
 currency hedging provides investors with the opportunity to potentially enhance the total
returns of their international holdings, through managing their currency exposure.
Page 504

The weakening outlook for Canada has many investors looking for opportunities
elsewhere, but investing internationally in individual foreign stocks can be challenging. Many
find buying and selling shares on markets outside of Canada difficult, confusing, and
expensive. ETFs, however, make it easy and cost-effective for investors, because they
provide a wide array of international exposures in a low-cost, transparent and convenient
way.
More and more investors are turning to ETFs to gain international exposures in the core of
their portfolios, where the strategy is to buy and hold for the long term. Cost-efficiency is
key. The good news is that with one trade, an investor can get access to a wide diversity of
stocks and geographies through an ETF. An example is the iShares Core MSCI All Country
World ex Canada Index ETF (XAW), which provides low-cost exposure to about 5,000
large-, mid- and small-cap companies in more than 40 countries, covering 99 percent of
global equity markets outside Canada. With XAW, investors can achieve near-total global
diversification, at one-tenth the cost of a traditional global equity mutual fund.
Beyond core holdings, however, ETFs are also increasingly finding favour among investors
who want to make more tactical plays on international markets to seek higher returns than
they can achieve through domestic allocations.
Whichever path investors choose, it’s clear that ETFs can provide new ways to realize the
benefits of portfolio diversification and growth potential beyond Canada’s borders.
Canadian investors really do have a world of opportunity to explore, and ETFs are helping
them get there.
Pat Chiefalo is the Managing Director, head of Canadian Product for iShares, BlackRock Asset Management Canada Limited. His comments are

excerpted with permission from “Going Global with ETFs,” Your Guide to ETF Investing (June 1, 2015). The information and opinions herein are provided for

informational purposes only, are subject to change and should not be relied upon as the basis for your investment decisions. © 2016 BlackRock. All rights

reserved. Used with the express permission of Pat Chiefalo.

XAMPLE 13.13
Risk and Return
Suppose the risk-free rate is 4%, the market risk premium is 8.6%, and a particular stock
has a beta of 1.3. Based on the CAPM, what is the expected return on this stock? What
would the expected return be if the beta were to double?
With a beta of 1.3, the risk premium for the stock would be 1.3 × 8.6% = 11.18%. The risk-
free rate is 4%, so the expected return is 15.18%. If the beta doubles to 2.6, the risk premium
would double to 22.36%, so the expected return would be 22.36% + 4% = 26.36%.

Concept Questions
. What is the fundamental relationship between risk and return in well-functioning markets?
. What is the security market line? Why must all assets plot directly on it in a well-functioning
market?
. What is the capital asset pricing model (CAPM)? What does it tell us about the required
return on a risky investment?
 LO4 13.8 | Arbitrage Pricing Theory and
Empirical Models
The CAPM and the arbitrage pricing theory (APT) are alternative models of risk and return. One advantage
of the APT is that it can handle multiple factors that the CAPM ignores. Although the bulk of our
presentation in this chapter focused on the one-factor model, a multifactor model is probably more
reflective of reality.
Page 505

The APT assumes that stock returns are generated according to factor models. For example, we have
described a stock’s return as:
Total return = Expected return + Unexpected return

R = E(R) + U

In APT, the unexpected return is related to several market factors. Suppose there are three such factors:
unanticipated changes in inflation, GNP, and interest rates. The total return can be expanded as:

R = E(R) + βI FI + βGN P FGN P + βrFr + ε [13.11]

The three factors FI , FGNP, and Fr represent systematic risk because these factors affect many securities.
The term ε is considered unsystematic risk because it is unique to each individual security.
Under this multifactor APT, we can generalize from three to K factors to express the relationship between
risk and return as:
E(R) = RF + E[(R1) − RF ]β1 + E[(R2) − RF ]β2 + E[(R3) − RF ]β3 + … + E[(RK ) − RF ]βK [13.12]

In this equation, β1 stands for the security’s beta with respect to the first factor, β2 stands for the security’s
beta with respect to the second factor, and so on. For example, if the first factor is inflation, β1 is the
security’s inflation beta. The term E(R1 ) is the expected return on a security (or portfolio) whose beta with
respect to the first factor is one and whose beta with respect to all other factors is zero. Because the market
compensates for risk, E[(R1 ) − Rf)] is positive in the normal case. 12 (An analogous interpretation can be
given to E(R2), E(R3 ), and so on.)
The equation states that the security’s expected return is related to its factor betas. The argument is that each
factor represents risk that cannot be diversified away. The higher a security’s beta is with regard to a
particular factor, the higher the risk that security bears. In a rational world, the expected return on the
security should compensate for this risk. The preceding equation states that the expected return is a
summation of the risk-free rate plus the compensation for each type of risk the security bears.
As an example, consider a Canadian study where the factors were:
1. The rate of growth in industrial production (INDUS).
2. The changes in the slope of the term structure of interest rates (the difference between the yield on long-
term and short-term Canada bonds) (TERMS).
3. The default risk premium for bonds (measured as the difference between the yield on long-term Canada
bonds and the yield on the ScotiaMcLeod corporate bond index) (RISKPREM).
4. The inflation (measured as the growth of the consumer price index) (INFL).
5. The value-weighted return on the market portfolio (S&P/TSX Composite) (MKRET). 13

Page 506

Using the period 1970–84, the empirical results of the study indicated that expected monthly returns on a
sample of 100 TSX stocks could be described as a function of the risk premiums associated with these five
factors.
Because many factors appear on the right side of the APT equation, the APT formulation explained
expected returns in this Canadian sample more accurately than did the CAPM. However, as we mentioned
earlier, one cannot easily determine which are the appropriate factors. The factors in this study were
included for reasons of both common sense and convenience. They were not derived from theory, and the
choice of factors varies from study to study. A more recent Canadian study, for example, includes changes
in a U.S. stock index and in exchange rates as factors. 14
The CAPM and the APT by no means exhaust the models and techniques used in practice to measure the
expected return on risky assets. Both the CAPM and the APT are risk based. They each measure the risk of a
security by its beta(s) on some systematic factor(s), and they each argue that the expected excess return
must be proportional to the beta(s). As we have seen, this is intuitively appealing and has a strong basis in
theory, but there are alternative approaches.
One popular alternative is a multifactor empirical model developed by Fama and French and based less on a
theory of how financial markets work and more on simply looking for regularities and relations in the past
history of market data. In such an approach, the researcher specifies some parameters or attributes
associated with the securities in question and then examines the data directly for a relation between these
attributes and expected returns. Fama and French examine whether the expected return on a firm is related to
its size and market to book ratio in addition to its beta. Is it true that small firms have higher average
returns than large firms? Do growth companies with high market to book ratios have higher average returns
than value companies with low market to book ratios?15 A well-known extension of the Fama-French
model includes a fourth, momentum, factor measured by last year’s stock return. 16
Although multifactor models are commonly used in investment performance analysis, they have not become
standard practice in estimating the cost of capital. Surveys of corporate executives show that only one in
three employ multifactor models for this purpose while over 70% rely on the CAPM. 17

Concept Questions
. What is the main advantage of the APT over the CAPM?
 Page 507

Summary and Conclusions

This chapter covered the essentials of risk. Along the way, we introduced a number of definitions
and concepts. The most important of these is the security market line, or SML. The SML is
important because it tells us the reward offered in financial markets for bearing risk. Once we know
this, we have a benchmark against which to compare the returns expected from real asset
investments and to determine if they are desirable.
Because we covered quite a bit of ground, it’s useful to summarize the basic economic logic
underlying the SML as follows:
1. Based on capital market history, there is a reward for bearing risk. This reward is the risk
premium on an asset.
2. The total risk associated with an asset has two parts: systematic risk and unsystematic risk.
Unsystematic risk can be freely eliminated by diversification (this is the principle of
diversification), so only systematic risk is rewarded. As a result, the risk premium on an asset is
determined by its systematic risk. This is the systematic risk principle.
3. An asset’s systematic risk, relative to average, can be measured by its beta coefficient, β.i The
risk premium on an asset is then given by its beta coefficient multiplied by the market risk
premium [E (R ) − R ] × β .
M f i

4. The expected return on an asset, E(Ri), is equal to the risk-free rate, Rf, plus the risk premium:
E(Ri) = Rf + [E(RM ) − Rf ] × βi

This is the equation of the SML, and it is often called the capital asset pricing model (CAPM).

This chapter completes our discussion of risk and return and concludes Part 5 of our book.
Now that we have a better understanding of what determines a firm’s cost of capital for an
investment, the next several chapters examine more closely how firms raise the long-term capital
needed for investment.
Key Terms
arbitrage pricing theory (APT)
beta coefficient
capital asset pricing model (CAPM)
expected return
market risk premium
portfolio
portfolio weights
principle of diversification
security market line (SML)
systematic risk
systematic risk principle
unsystematic risk
Page 508
Chapter Review Problems
and Self-Test
1. Expected Return and Standard Deviation This problem gives you some practice
calculating measures of prospective portfolio performance. There are two assets and three states
of the economy:

Rate of Return if State Occurs

State of E conomy Probability of State of E conomy Stock A Stock B
Recession .20 −.15 .20
Normal .50 .20 .30
Boom .30 .60 .40

What are the expected returns and standard deviations for these two stocks?
2. Portfolio Risk and Return Using the information in the previous problem, suppose you have
$20,000 total. If you put $15,000 in Stock A and the remainder in Stock B, what will be the
expected return and standard deviation on your portfolio?
3. Risk and Return Suppose you observe the following situation:

Security Beta Expected Return

Cooley Inc. 1.8 22.00%
Moyer Company 1.6 20.44

If the risk-free rate is 7%, are these securities correctly priced? What would the risk-free rate have
to be if they are correctly priced?
4. CAPM Suppose the risk-free rate is 8%. The expected return on the market is 16%. If a
particular stock has a beta of .7, what is its expected return based on the CAPM? If another
stock has an expected return of 24%, what must its beta be?
Answers to Self-Test Problems
1. The expected returns are just the possible returns multiplied by the associated probabilities:

E(RA) = (.20 × −.15) + (.50 × .20) + (.30 × .60) = 25%

E(RB) = (.20 × .20) + (.50 × .30) + (.30 × .40) = 31%

Page 509

The variances are given by the sums of the squared deviations from the expected returns
multiplied by their probabilities:
2 2 2 2
σ A
= .20 × (−.15 − .25) + .50 × (.20 − .25) + .30 × (.60 − .25)

2 2 2
= (.20 × −.40 ) + (.50 × −.05 ) + (.30 × .35 )

= (.20 × .16) + (.50 × .0025) + (.30 × .1225)

= .0700

2 2 2 2
σ B
= .20 × (.20 − .31) + .50 × (.30 − .31) + .30 × (.40 − .31)

2 2 2
= (.20 × −.11 ) + (.50 × −.01 ) + (.30 × .09 )

= (.20 × .0121) + (.50 × .0001) + (.30 × .0081)

= .0049

The standard deviations are thus:

σA = √ .0700 = 26.46%

σB = √ .0049 = 7%

2. The portfolio weights are $15,000/20,000 = .75 and $5,000/20,000 = .25. The expected
return is thus:
E(RP ) = .75 × E(RA) + .25 × E(RB)

= (.75 × 25%) + (.25 × 31%)

= 26.5%

Alternatively, we could calculate the portfolio’s return in each of the states:

State of Economy P robability of State of Economy P ortfolio Return if State Occurs

Recession .20 (.75 × −.15) + (.25 × .20) = −.0625
Normal .50 (.75 × .20) + (.25 × .30) = .2250
Boom .30 (.75 × .60) + (.25 × .40) = .5500

The portfolio’s expected return is:

E(RP ) = (.20 × −.0625) + (.50 × .2250) + (.30 × .5500) = 26.5%

This is the same as we had before.

The portfolio’s variance is:
2 2 2
 2 2 2 2
σ P
= .20 × (−.0625 − .265) + .50 × (.225 − .265) + .30 × (.55 − .265)

= .0466

So the standard deviation is √.0466 = 21.59%.

3. If we compute the reward-to-risk ratios, we get (22% − 7%)/1.8 = 8.33% for Cooley versus
8.4% for Moyer. Relative to that of Cooley, Moyer’s expected return is too high, so its price is
too low.
Page 510

If they are correctly priced, they must offer the same reward-to-risk ratio. The risk-free rate
would have to be such that:
(22% − Rf )/1.8 = (20.44% − Rf )/1.6

With a little algebra, we find that the risk-free rate must be 8%:

22% − Rf = (20.44% − Rf )(1.8/1.6)

22% − 20.44% × 1.125 = Rf − Rf × 1.125

Rf = 8%

4. Because the expected return on the market is 16%, the market risk premium is
16% − 8% = 8% . (the risk-free rate is 8%). The first stock has a beta of .7, so its expected

return is 8% + .7 × 8% = 13.6% .
For the second stock, notice that the risk premium is 24% − 8% = 16% . Because this is
twice as large as the market risk premium, the beta must be exactly equal to 2. We can verify this
using the CAPM:

E(Ri) = Rf + [E(RM ) − Rf ] × βi

24% = 8% + (16% − 8%) × β

βi = 16%/8%

= 2.0
Concepts Review
and Critical Thinking Questions
1. (LO3) In broad terms, why is some risk diversifiable? Why are some risks nondiversifiable?
Does it follow that an investor can control the level of unsystematic risk in a portfolio, but not
the level of systematic risk?
2. (LO3) Suppose the government announces that, based on a just-completed survey, the growth
rate in the economy is likely to be 2% in the coming year, as compared to 5% for the year just
completed. Will security prices increase, decrease, or stay the same following this
announcement? Does it make any difference whether or not the 2% figure was anticipated by the
market? Explain.
3. (LO3) Classify the following events as mostly systematic or mostly unsystematic. Is the
distinction clear in every case?
a. Short-term interest rates increase unexpectedly.
b. The interest rate a company pays on its short-term debt borrowing is increased by its bank.
c. Oil prices unexpectedly decline.
d. An oil tanker ruptures, creating a large oil spill.
e. A manufacturer loses a multimillion-dollar product liability suit.
f. A Supreme Court of Canada decision substantially broadens producer liability for injuries
suffered by product users.
Page 511

4. (LO3) Indicate whether the following events might cause stocks in general to change price, and
whether they might cause Big Widget Corp.’s stock to change price.
a. The government announces that inflation unexpectedly jumped by 2% last month.
b. Big Widget’s quarterly earnings report, just issued, generally fell in line with analysts’
expectations.
c. The government reports that economic growth last year was at 3%, which generally agreed
with most economists’ forecasts.
d. The directors of Big Widget die in a plane crash.
e. The Government of Canada approves changes to the tax code that will increase the top
marginal corporate tax rate. The legislation had been debated for the previous six months.
5. (LO1) If a portfolio has a positive investment in every asset, can the expected return on the
portfolio be greater than that on every asset in the portfolio? Can it be less than that on every
asset in the portfolio? If you answer yes to one or both of these questions, give an example to
support your answer.
6. (LO2) True or false: The most important characteristic in determining the variance of return of a
well-diversified portfolio is the variances of the individual assets in the portfolio. Explain.
 7. (LO2) If a portfolio has a positive investment in every asset, can the standard deviation on the
portfolio be less than that on every asset in the portfolio? What about the portfolio beta?
8. (LO4) Is it possible that a risky asset could have a beta of zero? Explain. Based on the CAPM,
what is the expected return on such an asset? Is it possible that a risky asset could have a
negative beta? What does the CAPM predict about the expected return on such an asset? Can
you give an explanation for your answer?
9. (LO4) In our discussion of the SML, we defined alpha. What does alpha measure? What alpha
would you like to see on your investments?
10. (LO4) Common advice on Wall Street is “Keep your alpha high and your beta low.” Why?
11. (LO1) In recent years, it has been common for companies to experience significant stock price
changes in reaction to announcements of massive layoffs. Critics charge that such events
encourage companies to fire long-time employees and that Bay Street is cheering them on. Do
you agree or disagree?
12. (LO1) As indicated by a number of examples in this chapter, earnings announcements by
companies are closely followed by, and frequently result in, share price revisions. Two issues
should come to mind. First, earnings announcements concern past periods. If the market values
stocks based on expectations of the future, why are numbers summarizing past performance
relevant? Second, these announcements concern accounting earnings. Going back to
Chapter 2, such earnings may have little to do with cash flow, so, again, why are they
relevant?
 Page 512

Questions and Problems

Basic (Questions 1–20)

1. Determining Portfolio Weights (LO1) What are the portfolio weights for a portfolio that
has 135 shares of Stock A that sell for $48 per share and 165 shares of Stock B that sell for
$29 per share?
2. Portfolio Expected Return (LO1) You own a portfolio that has $2,650 invested in Stock A
and $4,450 invested in Stock B. If the expected returns on these stocks are 8% and 11%,
respectively, what is the expected return on the portfolio?
3. Portfolio Expected Return (LO1) You own a portfolio that is 35% invested in Stock X,
20% in Stock Y, and 45% in Stock Z. The expected returns on these three stocks are 9%,
17%, and 13%, respectively. What is the expected return on the portfolio?
4. Portfolio Expected Return (LO1) You have $10,000 to invest in a stock portfolio. Your
choices are Stock X with an expected return of 11.5% and Stock Y with an expected return
of 9.4%. If your goal is to create a portfolio with an expected return of 10.85%, how much
money will you invest in Stock X? In Stock Y?
5. Calculating Expected Return (LO1) Based on the following information, calculate the
expected return:

State of Economy P robability of State of Economy P ortfolio Return if State Occurs

Recession .20 −.14
Boom .80 .17

6. Calculating Expected Return (LO1) Based on the following information, calculate the
expected return:

State of Economy P robability of State of Economy P ortfolio Return if State Occurs

Recession .10 −.18
Normal .60 .11
Boom .30 .26

7. Calculating Returns and Standard Deviations (LO1) Based on the following information,
calculate the expected return and standard deviation for the two stocks:

Rate of Return if State Occurs

State of Economy P robability of State of Economy Stock A Stock B
Recession .15 .04 −.17
Normal .55 .09 .12
Boom .30 .17 .27
 8. Calculating Expected Returns (LO1) A portfolio is invested 25% in Stock G, 55% in Stock
J, and 20% in Stock K. The expected returns on these stocks are 8%, 14%, and 18%,
respectively. What is the portfolio’s expected return? How do you interpret your answer?
Page 513
9. Returns and Variances (LO1, 2) Consider the following information:

Rate of Return if State Occurs

State of Economy P robability of State of Economy Stock A Stock B Stock c
Boom .65 .07 .15 .33
Bust .35 .13 .03 −.06

a. What is the expected return on an equally weighted portfolio of these three stocks?

b. What is the variance of a portfolio invested 20% each in A and B and 60% in C?

10. Returns and Standard Deviations (LO1, 2) Consider the following information:

Rate of Return if State Occurs

State of Economy P robability of State of Economy Stock A Stock B Stock C
Boom .10 .35 .45 .27
Good .60 .16 .10 .08
Poor .25 −.01 −.06 −.04
Bust .05 −.12 −.20 −.09

a. Your portfolio is invested 30% each in A and C, and 40% in B. What is the expected
return of the portfolio?

b. What is the variance of this portfolio? The standard deviation?

11. Calculating Portfolio Betas (LO4) You own a stock portfolio invested 20% in Stock Q,
30% in Stock R, 35% in Stock S, and 15% in Stock T. The betas for these four stocks are
.84, 1.17, 1.08, and 1.36, respectively. What is the portfolio beta?
12. Calculating Portfolio Betas (LO4) You own a portfolio equally invested in a risk-free asset
and two stocks. If one of the stocks has a beta of 1.32 and the total portfolio is equally as
risky as the market, what must the beta be for the other stock in your portfolio?
13. Using CAPM (LO1, 4) A stock has a beta of 1.15, the expected return on the market is
10.3%, and the risk-free rate is 3.8%. What must the expected return on this stock be?
14. Using CAPM (LO1, 4) A stock has an expected return of 10.2%, the risk-free rate is 4.1%,
and the market risk premium is 7.2%. What must the beta of this stock be?
15. Using CAPM (LO1, 4) A stock has an expected return of 11.05%, its beta is 1.13, and the
risk-free rate is 3.6%. What must the expected return on the market be?
16. Using CAPM (LO4) A stock has an expected return of 12.15%, its beta is 1.31, and the
expected return on the market is 10.2%. What must the risk-free rate be?
Page 514
17. Using CAPM (LO1, 4) A stock has a beta of 1.14 and an expected return of 10.5%. A risk-
 free asset currently earns 2.4%.
a. What is the expected return on a portfolio that is equally invested in the two assets?

b. If a portfolio of the two assets has a beta of .92, what are the portfolio weights?

c. If a portfolio of the two assets has an expected return of 9%, what is its beta?

d. If a portfolio of the two assets has a beta of 2.28, what are the portfolio weights? How
do you interpret the weights for the two assets in this case? Explain.

18. Using the SML (LO1, 4) Asset W has an expected return of 11.8% and a beta of 1.15. If the
risk-free rate is 3.7%, complete the following table for portfolios of Asset W and a risk-free
asset. Illustrate the relationship between portfolio expected return and portfolio beta by
plotting the expected returns against the betas. What is the slope of the line that results?

P ercentage of P ortfolio in Asset W P ortfolio Expected Return P ortfolio Beta

0%
25
50
75
100
125
150

19. Reward-to-Risk Ratios (LO4) Stock Y has a beta of 1.2 and an expected return of 11.4%.
Stock Z has a beta of .80 and an expected return of 8.06%. If the risk-free rate is 2.5% and
the market risk premium is 7.2%, are these stocks correctly priced?
20. Reward-to-Risk Ratios (LO4) In the previous problem, what would the risk-free rate have to
be for the two stocks to be correctly priced?

Intermediate (Questions 21–24)

21. Portfolio Returns (LO1, 2) Using Table 12.4 from the previous chapter on capital
market history, determine the return on a portfolio that is equally invested in Canadian large-
company stocks and long-term government bonds. What is the return on a portfolio that is
equally invested in small-company stocks and Treasury bills?
22. CAPM (LO4) Using the CAPM, show that the ratio of the risk premiums on two assets is
equal to the ratio of their betas.
23. Portfolio Returns and Deviations (LO1, 2) Consider the following information about
three stocks:

Rate of Return if State Occurs

State of Economy P robability of State of Economy Stock A Stock B Stock C
Boom .20 .24 .36 .55
Normal .55 .17 .13 .09
 Rate of Return if State Occurs
Bust .25 .00 −.28 −.45

a. If your portfolio is invested 40% each in A and B and 20% in C, what is the portfolio
expected return? The variance? The standard deviation?

b. If the expected T-bill rate is 3.80%, what is the expected risk premium on the portfolio?

c. If the expected inflation rate is 3.50%, what are the approximate and exact expected real
returns on the portfolio? What are the approximate and exact expected real risk
premiums on the portfolio?
Page 515

24. Analyzing a Portfolio (LO2) You want to create a portfolio equally as risky as the market,
and you have $1,000,000 to invest. Given this information, fill in the rest of the following
table:

Asset Inv estment Beta

Stock A $185,000 .80
Stock B $320,000 1.13
Stock C 1.29
Risk-free asset

Challenge (Questions 25–31)

25. Analyzing a Portfolio (LO2, 4) You have $100,000 to invest in a portfolio containing
Stock X and Stock Y. Your goal is to create a portfolio that has an expected return of
13.6%. If Stock X has an expected return of 11.4% and a beta of 1.25, and Stock Y has an
expected return of 8.68% and a beta of .85, how much money will you invest in stock Y?
How do you interpret your answer? What is the beta of your portfolio?
26. Systematic versus Unsystematic Risk (LO3) Consider the following information about
Stocks I and II:

Rate of Return if State Occurs

State of Economy P robability of State of Economy Stock I Stock II
Recession .15 .02 −.25
Normal .70 .21 .09
Irrational .15 .06 .44
exuberance

The market risk premium is 7%, and the risk-free rate is 4%. Which stock has the most
systematic risk? Which one has the most unsystematic risk? Which stock is “riskier”?
Explain.
27. SML (LO4) Suppose you observe the following situation:
 Security Beta Expected Return
Pete Corp. 1.15 .129
Repete Co. .84 .102

Page 516

Assume these securities are correctly priced. Based on the CAPM, what is the expected
return on the market? What is the risk-free rate?
28. SML (LO1, 4) Suppose you observe the following situation:

Return if State Occurs

State of Economy P robability of State Stock A Stock B
Recession .25 −.08 −.05
Normal .60 .13 .14
Irrational exuberance .15 .48 .29

a. Calculate the expected return on each stock.

b. Assuming the capital asset pricing model holds and stock A’s beta is greater than stock
B’s beta by .25, what is the expected market risk premium?

29. Using CAPM (LO1, 4) A portfolio invests in a risk-free asset and the market portfolio has
an expected return of 7% and a standard deviation of 10%. Suppose the risk-free rate is 4%,
and the standard deviation on the market portfolio is 22%. According to the CAPM, what
expected rate of return would a security earn if it had a 0.55 beta?
30. Using CAPM (LO1, 4) A security’s beta can be calculated using covariance of the
security’s return with the market, divided by variance of return of the market portfolio. We
will show why this is true in Appendix 13A. Suppose the risk-free rate is 4.8% and the
market portfolio has an expected return of 11.4%. The market portfolio has a variance of
0.0429. Portfolio Z has a covariance of 0.39 with the market portfolio. According to the
CAPM, what is the expected return on portfolio Z?
31. Using CAPM (LO1, 4) There are two stocks in the market, stock A and stock B. The price
of stock A today is $65. The price of stock A next year will be $53 if the economy is in a
recession, $73 if the economy is normal, and $85 if the economy is expanding. The
probabilities of recession, normal times, and expansion are 0.2, 0.6, and 0.2, respectively.
Stock A pays no dividends and has a beta of 0.68. Stock B has an expected return of 13%, a
standard deviation of 34%, a beta of 0.45, and a correlation with stock A of 0.48. The
market portfolio has a standard deviation of 14%. Assume the CAPM holds.
a. What are the expected return and standard deviation of stock A?

b. If you are a typical, risk-averse investor with a well-diversified portfolio, which stock
would you prefer? Why?

c. What are the expected return and standard deviation of a portfolio consisting of 60% of
stock A and 40% of stock B?

d. What is the beta of the portfolio in (c)?

 Page 517

INTERNET APPLICATION QUESTIONS

1. You have decided to invest in an equally weighted portfolio consisting of Petro-Canada,
Royal Bank of Canada, Canadian Tire, and Air Canada and need to find the beta of your
portfolio. Go to finance.yahoo.com and find the ticker symbols for each of these
companies. Next, go back to finance.yahoo.com, enter one of the ticker symbols and get
a stock quote. Find the beta for each of the companies. What is the beta for your
portfolio?
2. Go to finance.yahoo.com, search for Aritzia (ATZ), and find the beta for the company.
Go to bankofcanada.ca/rates/interest-rates and find the current interest rate for three-
month Treasury bills. Using this information and 7% as an estimated return of the market,
calculate the expected return for ATZ using the CAPM. What would be the expected
stock price one year from now?
3. Recall that the site theglobeandmail.com/globe-investor/funds-and-etfs/funds contains
considerable information on Canadian mutual funds. Visit the site and update the
calculations in Example 13.12 to reflect the most recent three-year period.
4. You want to find the expected return for Bank of Montreal using the CAPM. First, you
need the market risk premium. Go to bmonesbittburns.com/economics, and find Daily
Economic Update under Publications. Find the current interest rate for three-month
Treasury bills. Use the average Canadian common stock return in Table 12.3 to
calculate the market risk premium. If the beta for Bank of Montreal is 1.01, what is the
expected return using CAPM?18 What assumptions have you made to arrive at this
number? As you may recall from Chapter 8, stock growth is often assumed to be
equal to earnings growth. Compare your answer above with an EPS growth estimate from
theglobeandmail.com/globe-investor. What does this tell you about analyst estimates?
5. You have decided to invest in an equally weighted portfolio consisting of Rogers
Communications, Bank of Montreal, and Goldcorp Inc. and need to find the beta of your
portfolio. Go to finance.yahoo.com and follow the Symbol Lookup link to find the ticker
symbols for each of these companies. Next, go back to finance.yahoo.com, enter one of
the ticker symbols and get a stock quote. Follow the Profile link to find the beta for this
company. You will then need to find the beta for each of the companies. What is the beta
for your portfolio? (Note that this beta will compare the stock to the NYSE.)
6. Go to finance.yahoo.com and enter the ticker symbol RCI for Rogers Communication
Inc. Follow the Profile link to get the beta for the company. Next, follow the Research
link to find the estimated price in 12 months according to market analysts. Using the
current share price and the mean target price, compute the expected return for this stock.
Don’t forget to include the expected dividend payments over the next year. Now go to ba
nkofcanada.ca/rates/interest-rates and find the current interest rate for three-month
Treasury bills. Using this information, calculate the expected return on the market using
the reward-to-risk ratio. Does this number make sense? Why or why not? (Note that the
beta value you locate will compare Rogers Communication to the NYSE volatility. You
should analyze this question from a U.S. perspective.)
 Page 518

APPENDIX 13A

Derivation of the Capital Asset Pricing Model

Up to this point, we have assumed that all assets on the efficient frontier are risky. Alternatively, an
investor could easily combine a risky investment with an investment in a riskless or risk-free security,
such as a Canada Treasury bill. Using the equation for portfolio variance ( Equation 13A.1) we can
find the variance of a portfolio with one risky and one risk-free asset:

2 2 2 2 2
[13A.1]
σ P = x Lσ L + x U σ U + 2x L x U CORR L,U σ L σ U

However, by definition, the risk-free asset (say, L in this example) has no variability so the equation for
portfolio standard deviation reduces to:
2 2 2
σ p = x U σ U

2
σ p = √σ p = x U σ U

The relationship between risk and return for one risky and one riskless asset is represented on a straight
line between the risk-free rate and a pure investment in the risky asset as shown in Figure 13A.1. The
line extends to the right of the point representing the risky asset when we assume the investor can borrow
at the risk-free rate to take a leveraged position of more than 100% in the risky asset.
FIGURE 13A.1
Relationship between expected return and standard dev iation for an inv estment in a combination of risky securities and the
riskless asset
To form an optimal portfolio, an investor is likely to combine an investment in the riskless asset with a
portfolio of risky assets. Figure 13A.1 illustrates our discussion by showing a risk-free asset and four
risky assets: A, X, Q, and Y. If there is no riskless asset, the efficient set is the curve from X to Y. With a
risk-free asset, it is possible to form portfolios like 1, 2, and 3 combining Q with the risk-free asset.
Portfolios 4 and 5 combine the riskless asset with A.
Page 519

The graph illustrates an important point. With riskless borrowing and lending, the portfolio of risky
assets held by any investor would always be point A. Regardless of the investor’s tolerance for risk, they
would never choose any other point on the efficient set of risky assets. Rather, an investor with a high
aversion to risk would combine the securities of A with riskless assets. The investor would borrow at the
risk-free rate to invest more funds in A had they low aversion to risk. In other words, all investors would
choose portfolios along Line II, called the capital market line.
To move from our description of a single investor to market equilibrium, financial economists imagine a
world where all investors possess the same estimates of expected returns, variance, and correlations.
This assumption is called homogeneous expectations.

If all investors have homogeneous expectations, Figure 13A.1 becomes the same for all individuals.
All investors sketch out the same efficient set of risky assets because they are working with the same
inputs. This efficient set of risky assets is represented by the curve XAY. Because the same risk-free rate
applies to everyone, all investors view point A as the portfolio of risky assets to be held. In a world with
homogeneous expectations, all investors would hold the portfolio of risky assets represented by point A.

19
If all investors choose the same portfolio of risky assets, A, then A must be the market portfolio. 19 This
is because, in our simplified world of homogeneous expectations, no asset would be demanded (and
priced) if it were not in portfolio A. Since all assets have some demand and non-zero price, A has to be
the market portfolio including all assets.
The variance of the market portfolio can be represented as:

x i x j σ ij
[13A.2]

where we define σij as the covariance of i with j if i ≠ j and σij is the variance or σ2i if i = j.
σ ij = CORR i,j σ i σ j

Using a little elementary calculus, we can represent a security’s systematic risk (the contribution of
security i to the risk of the market portfolio) by taking the partial derivative of the portfolio risk with
respect to a change in the weight of the security. This measures the change in the portfolio variance when
the weight of the security is increased slightly. For security 2,

N
[13A.3]
2
δσ p
2
= 2 ∑ x j σ i2 = 2 [x 1 COV (R 1 , R 2 ) + x 2 σ 2 + x 3 COV (R 3 , R 2 ) + … + x N COV (R N , R 2 )]
δx 2
j=1

The term within brackets (in Equation 13A.3) is COV(R2, RM). This shows that systematic risk is
proportional to a security’s covariance with the market portfolio.
The final step is to standardize systematic risk by dividing by the variance of the market portfolio. The
result is β2 as presented in the text.
[13A.4]
COV(R 2 , R M )
β2 =
2
σ (R M )

If you consult any basic statistics text, you will see that this formula is identical to the β2 obtained from
a regression of R2 on RM.

We can now redraw Figure 13A.1 in expected return-β space, as shown in Figure 13A.2. The
vertical axis remains the same, but on the horizontal axis we replace total risk (σ) with systematic risk as
measured by β. We plot the two points on the capital market line from Figure 13A.1: RF with β = 0
and M (the market portfolio represented by A) with a β = 1. To see that β = 1, substitute portfolio M

M for i in Equation 13A.4.

COV(R M , R M )
βM =
2
σ (R M )

CORR M ,M σ M σ M
=
2
σ (R M )

2
1.0 × σ
M
=
2
σ (R M )

βM = 1.0

Page 520
The result is the security market line shown in Figure 13A.2. We can use the slope-intercept method
to find that the intercept of the SML is RF and the slope is (RM−RF). The equation for the SML is:
E(R) = R F + β(R M − R F )

And this completes the derivation of the capital asset pricing model.
FIGURE 13A.2
Relationship between expected return on an indiv idual security and beta of the security

Appendix Question and Problem

A.1 A mutual fund A has a standard deviation of 13% (assume this fund to be on the efficient
frontier; that is, the fund plots on the capital market line). The risk-free rate is 3%. The standard
deviation of the market’s return is 18%, and the expected return on the market is 15%. What is the
expected return on the mutual fund A?