Chapter 8
Chapter 8
RiskandReturn
INTRODUCTION
Chapter 7 introduced risk and risk measurement. The chapter also showed us how diversification
enables an investor to reduce risk, but that there is a limit to risk reduction. Only risk unique to
individual securities can be diversified away; the risk that a security bears as part of the overall
economic system remains in the portfolio. This risk was termed market risk. This chapter builds
on the lessons of chapter 7 and presents formal models linking risk and expected return. The
chapter begins with a discussion on portfolio theory pioneered by Harry Markowitz. This theory
formed the basis for the most popular and widely used model of risk-return relationship: the
Capital Asset Pricing Model or CAPM. The CAPM has some empirical evidence to support the
basic conclusions derived from the model. Recently, it has been challenged by other theories that
purport to rely on assumptions less restrictive than the ones used by CAPM. The chapter presents
three alternate theories to CAPM. While CAPM is far from perfect, it remains the most widely
accepted model, primarily because of its relative simplicity and ease of application.
KEY CONCEPTS IN THE CHAPTER
Portfolio Theory: Stock market returns of individual stocks covering relatively short periods
closely approximate the normal distribution. This enables one to describe the entire distribution of
the returns in terms of two parameters- the mean and standard deviation. Individual investors are
assumed to be risk-averse, meaning that they would like to be compensated for bearing more risk.
This also means that they like higher returns and lower risk. The standard deviation of returns
measures risk.
The portfolio theory gives the basic principles of selecting the optimal portfolio in terms of risk
and return. We learned in the last chapter that when two stocks are combined into a portfolio, the
risk (standard deviation of the portfolio) is less than the average of the two individual standard
deviations. Of course, when combining securities into a portfolio, an infinite number of
combinations are possible. But investors would choose only those which are efficient portfolios.
Portfolios are considered efficient if, for a given standard deviation, they give the highest return,
or for a given return they have the lowest standard deviation. Investors need only consider
efficient portfolios, for all other portfolios give them a poor deal.
Investors have more than two stocks to choose and we can extend the idea of efficient portfolio to
the universe of available securities. All the efficient portfolios can be identified and depending on
ones own risk-return preference, an investor can choose the best portfolio of his liking. The
technique of quadratic programming can be used to choose the efficient portfolio of ones
preference. If one were to graph the set of all efficient portfolios selected from all the stocks
available it will look like the curve shown in Figure 1. Every investor will only choose a portfolio
on this curve. Any choice not on the efficient set will be dominated by a portfolio on the efficient
83
B
A
Standard Deviation
Investors choices can be expanded by introducing the possibility of lending or borrowing at the
risk-free rate. This will allow the investors to combine investing in the stock portfolios given by
the efficient set with lending or borrowing (borrowing can be seen as negative lending). The
possibilities are shown in Figure 2.
Figure 2
S
Expected
Return
rf
B
A
Standard Deviation
The figure shows that investors can choose along the line starting from the risk-free rate to the
portfolio S. If the investor chooses to invest all her money in risk-free lending, she will have the
risk-free rate at zero standard deviation. Any point between rf and S will have part of the money
invested in the stock portfolio S and the remaining lent at the risk-free rate. Any investor is free
to choose points to the right of S. This would mean that the investor is borrowing money at the
risk free rate and investing that along with his own money in portfolio S. You can clearly see that
regardless of where the investor wants to be, the stock portfolio will be the same, namely the
portfolio S. In other words, S is the best of all the efficient portfolios and every investor will be
invested in this portfolio. This will also mean that every stock will be held as part of this
portfolio.
An investor's job can be separated into two parts: first, to choose the best (risky) portfolio of
stocks S and, second, to combine it with the right amount of lending or borrowing to adjust to the
preferred level of risk. This is known as the separation theorem: the job of selecting individual
stocks in ones portfolio can be independently of ones risk preference. In a competitive market,
there is no reason to concentrate portfolios in particular stocks, and we can identify S as the
84
market portfolio. Everyone holds S. Risk preference only decides what fraction of ones
investment goes to S.
Capital Asset Pricing Model: Capital Asset Pricing Model (CAPM) is an extension of what we
saw in Figure 2. Investing in risky assets such as the market portfolio, S, should carry a premium
compared to the risk-free rate. Otherwise, investors will not take the risk. The market portfolio
is the basic benchmark of risk in the CAPM. The risk premium on the market portfolio, measured
as the difference between the market return and the risk-free rate is called the market risk
premium. The beta for the market portfolio is 1. The CAPM postulates a linear relationship
between risk and return as shown in Figure 3. Figure 3 shows the security market line (SML).
Figure 3
Returns
SMLS
rm
rf
0.5
1.0
Beta
The risk of each stock is measured by its beta and the risk premium varies in direct proportion to
beta. All stocks will lie along the SML and the expected return on a stock can be calculated by
adding the risk premium to the risk-free rate.
Expected risk premium = r r f = ( rm r f )
A key implication of the CAPM is that it is the risk of a stock that matters in its contribution to
portfolio risk. This is what is measured by beta and it depends on the stocks sensitivity to
changes in the value of the market portfolio.
Validity of CAPM: Over the years, a lot of research has been done to test the validity of CAPM.
Remember, however, the CAPM is a theory about expected returns. We can only measure actual
returns. This makes it difficult to test the theory as it is conceived. Another problem in testing
CAPM is that the market portfolio should include all assets, not just stocks traded in stock
exchanges. In practice, most of the tests use stock market indexes such as the S & P 500 as
proxies for the market portfolio.
One of the best known tests of the CAPMs validity involved analyzing portfolio returns over the
period 1931 to 1991. The study, conducted by Fischer Black, grouped all New York Stock
Exchange stocks into 10 portfolios selected according to the ranking of the betas of their stocks.
Every year these portfolios were readjusted, using monthly data from the previous 5 years, so that
85
portfolio 1 always contained the 10 percent of stocks with the lowest betas, portfolio 9 the next
10 percent, up to portfolio 10 with the highest beta stocks. The performance of these portfolios
was assessed over the 60-year period. The results showed a strong positive relationship between
beta and returns and thus broad support for the CAPM. However, it was found that the security
market line was flatter than the theory predicted and it was above the risk-free rate.
In recent years, the relationship between beta and the risk premium has been particularly flat. It
has been argued that some of these results could be attributed to sampling errors (the realized
average returns are not the same as the previously expected ones). There is also evidence that
suggests that beta is not the only factor that affects expected returns. Firms, which are small, and
ones, which have low market-to-book ratios, seem to provide higher average returns irrespective
of their betas. It is difficult to judge the economic significance of these results. The small-firm
effect is found to be less in more recent studies. It is also possible that some of the results which
cast doubts on the validity of the CAPM can be produced by statistical strategies (known as "data
mining"). The supporters of the CAPM believe that the broad conclusions of the theory are valid.
Consumption CAPM: The consumption CAPM is an alternate theory to the CAPM. The CAPM
assumes that investors are concerned with their wealth and uncertainty about the same. The
consumption CAPM holds that risk and expected return should be related to changes in investors
aggregate consumption rather than their wealth. Investors wealth is not directly relevant to the
stock returns. One need not worry about defining the market portfolio. However, you have to
estimate aggregate consumption and its changes. Empirically, it is observed that estimated
aggregate consumption changes rather gradually through time. This does not give a good
explanation of past average rates of return. In short, the model has not caught on and its practical
appeal seems limited.
Arbitrage Pricing Theory: The arbitrage pricing theory (APT) was the first serious challenger to
CAPM. It does not involve many of the restrictive assumptions of CAPM. It postulates that each
stock's return depends on several pervasive macroeconomic factors and the risk premium depends
on the factor weights. The general model can be written as:
Return = a + b1(factor 1) + b2(factor 2) +
Alternately, the risk premium: r - rf = b1(premium1) + b2 (premium2) +
The APT is general and has appealing features. The market portfolio has a role in the model and
thus we need not be concerned with measuring the same. The model is general and is dependent
on the absence of arbitrage opportunities. The model does not specify either the number of
factors or identity of these factors. The major problem is that there is little agreement on what the
factors are. There have been many empirical attempts at identifying the factors. These have been
met with limited success. The APT is still a long way from replacing the CAPM as the primary
theory for risk-return relationship.
The Three-Factor Model: Eugene Fama and Kenneth French incorporate the small-firm effect
and the book-to-market ratio effect seen in empirical studies into a formal risk-return relationship
model. The model can be represented as:
86
17 5
d. Required rate of return The investment return is less than the required
rate of return. Hence, its NPV will be negative.
3. ABC Corp. wishes to use the Fama-French 3-factor model to estimate its cost of equity
capital. The company has estimated the following factor sensitivities and factor risk
premiums:
Factors
Factor
Factor risk
87
Market factor
Size factor
Book-tomarket factor
Sensitivity
1.5
-0.4
premium
5.2 percent
3.2
0.2
5.4
If the risk free rate is 6 percent, estimate the cost of equity capital for the firm.
SOLUTION
Risk premium = 15
. x5.2 0.4 x 3.2 0.2 x5.4 7.60 percent
Cost of equity capital = 6 + 7.6 = 13.60 percent.
SUMMARY
The chapter describes the basic principles of portfolio selection and the theories of which provide
formal models of the linkage between risk and return. Risk averse investors choose portfolios
based on risk and return. It makes sense to choose only efficient portfolios--portfolios which
provide the highest return for a given level of risk. In presence of the possibility of risk-free
lending and borrowing, investors will end up choosing the one portfolio that, in combination with
risk-free lending/borrowing, gives the best choices regardless of ones own risk preferences. This
forms the basis for the widely accepted theory, Capital Asset Pricing Model. The CAPM uses the
market portfolio as the basic benchmark for risky investments and the risk of any stock or
investment is measured relative to the market portfolio. The risk measure is beta and it reflects
the stocks sensitivity to changes in the value of the market portfolio. The risk-return relationship
is captured by the security market line and the risk premium on any stock is proportional to its
beta.
The CAPM theory finds some empirical support, though recently questions have been raised
about the strict validity of the security market line. Alternate theories of risk-return relationship
include the consumption CAPM, the arbitrage pricing theory, and the three-factor model. Each of
these theories has intellectual appeal and might someday prove to be a replacement to the CAPM;
however for the present the CAPM still rules. The primary attraction of the CAPM is its relative
simplicity and ease of practical application. The CAPM should be seen as an approximate and
partial explanation of the risk-return trade off faced by the investor. We will find many
applications of the CAPM in the later chapters.
LIST OF TERMS
Arbitrage pricing theory
Capital Asset Pricing Model
Consumption beta
Consumption CAPM
Efficient portfolios
Expected risk premium
EXERCISES
88
Fill-in Questions
1. The ________________ uses less restrictive assumptions than the CAPM does and postulates
that returns are dependent on pervasive macroeconomic factors.
2. The distribution of stock returns when considered over fairly short periods, closely follow
_______________.
3. A(n) ____________ is a portfolio that gives the highest expected return for a given standard
deviation or the lowest standard deviation for a given return.
4. Fama and French proposed the ____________.
5. Optimal portfolio selection can be done using the________________ technique.
6. The _____________ relates the risk premium to the sensitivity of the investment to changes in
consumption, and this sensitivity measure is called the ___________.
7. The _______________ relates the risk premium to sensitivity of the assets returns to changes
in the value of the market portfolio.
8. The expected return on a stock investment, according to CAPM, equals the risk free rate plus
_______________.
9. The difference between the expected market return and the risk-free rate is _____________.
10. According to the CAPM, the expected returns from all investments must plot along the
_____________.
Problems
1. Pick the better investment from each of the following pairs of portfolios.
a. Portfolio A: expected return 14 percent, variance 400.
Portfolio B: expected return 13 percent, variance 441.
b. Portfolio J: expected return 20 percent, variance 529.
Portfolio K: expected return 20 percent, variance 400.
c. Portfolio R: expected return 8 percent, variance 225.
Portfolio S: expected return 9 percent, variance 225.
d. Portfolio X: expected return 12 percent, variance 380.
Portfolio Y: expected return 15 percent, variance 460.
2. Calculate the expected return for the following portfolios:
a. beta = 1.2
b. beta = 1.8
c. beta = 0.5
d. beta = 0
89
Assume the risk free rate to be 6 percent and the expected market risk premium to equal 15
percent.
3. Your broker is urging you to invest in one of three portfolios on which the returns are
expected to be as follows: portfolio A, 12 percent; portfolio B, 16 percent; portfolio C, 20
percent. You believe these estimates, but you also have sufficient data to calculate the betas of
the portfolios with confidence. You find the betas are 0.5 for A, 1.1 for B, and 2.0 for C.
Which portfolio is best and why? (Hint: See if you can duplicate portfolio B by some
combination of A and C.)
4. Stock J has a beta of 1.2 and an expected return of 15.6 percent, and stock K has a beta of 0.8
and an expected return of 12.4 percent. What must be (a) the expected return on the market
and (b) the risk-free rate of return, to be consistent with the capital asset pricing model?
5. Stock A has an expected return of 14 percent and a beta of 1.1. Stock B has an expected
return of 18. The risk free rate is 5.2 percent. Assume that stock A is correctly priced (i.e. it
is on the security market line). If stock B has a beta of 1.4, is it correctly priced?
6. An alternate form of the CAPM is written as: r = rm. Write a general expression for
and find the value for for each of the following cases: (a) beta = 1, (b) beta = 2, and (c)
beta = 0. What is the value of beta if = zero?
7. The return on a proposed investment is expected to be 24 percent. The standard deviation is
estimated at 24 percent. The risk free rate is 6 percent. Assume the expected market risk
premium to be 9 percent. Assume that the standard deviation of market returns to be 15
percent. If the project correlates fairly highly with the economy, comment on the
attractiveness of the project.
8. You are told that MM Inc. stock has a beta of 1.3. Estimate its cost of equity capital.
9. A three factor APT model has the following factors and risk premiums:
Factors
Risk premiums
Change in GNP
4%
Change in dollar exchange rates
-2
Change in energy prices
-1
Calculate the expected rates of return on the following stocks, assuming the risk-free rate to be
5.5 percent.
a. Company A: b1 = 0.5, b2 = -1.5, b3 = 0.3
b. Company B: b1 = 1.2, b2 = 0, b3 = -0.5
c. Company C: b1 = 2, b2 = 0.5, b3 = -2
10. Use the information on factor risk premiums from the worked example three-factor model
problem to calculate the required rate of return on the following stocks with factor
sensitivities to market, size, and book-to-market as follows:
a. Factor sensitivities: 1.2, 2, and 0
b. Factor sensitivities: 1.5, 0.3, and 1
90
Arbitragepricingtheory
Normaldistribution
Efficientportfolio
Thethreefactormodel
Quadraticprogramming
6.ConsumptionCAPM,consumptionbeta
7.Capitalassetpricingmodel
8.Riskpremium
9.Marketriskpremium
10.Securitymarketline
Problems
1. Theefficientportfoliosare:
a. PortfolioAhigherreturn,lowervariance
b. PortfolioKsamereturn,lowervariance
c. PortfolioShigherreturn,samevariance
d. Cannotdecidebetweenthetwowithoutknowingtheinvestorsriskpreference
2. Theexpectedreturnscanbecalculatedbytheformula:r=rf + (rm -rf)
a. r=6+1.2(9)=16.8%
b. r=22.2%
c. r=10.5%
d. r=6%
3. Aportfoliocontaining60percentAand40percentCwillhavethesamebetaasthatofBbut
haslowerreturn(15.2%)thanthatofB.Thus,Bisabetterchoice.
4. rf + 1.2(rm -rf) = 16, rf + 0.8(rm -rf) = 12.4 Solving for rf and rm , gives you: rf = 6 percent,
rm = 14 percent.
91
mp p
m
92