Principles of Finance | Topic 5: Risk and Return | Slide 1/54

Principles of Finance

Topic 5: Risk and Return

Dr Yue (Lucy) Liu

University of Edinburgh Business School

Slides are based on Berk and DeMarzo (3rd edition), Chapters 10 & 11
Principles of Finance | Topic 5: Risk and Return | Slide 2/54

Synopsis

Last time, we learned how to price options.

We learned that option contracts give their owners certain rights but no
obligations.

We used binomial and Black-Scholes option pricing models to price

options.

This time, we will investigate risk and return of securities.

Principles of Finance | Topic 5: Risk and Return | Slide 3/54

Motivation for this lecture

Let’s recall how we priced bonds and stocks.

The price of a bond is the present value of its certain cash

flows (i.e., coupons and face value), where the discount
rate is the risk-free rate rf .

The price of a stock is the present value of its expected cash flows (i.e.,
dividends and selling price), where the discount rate is a risk-adjusted
rate re .

Note that re > rf , and the difference is called the risk premium. Risk
premium is a consequence of investors’ risk aversion.

In this lecture, we address the following question: How do we calculate

the risk-adjusted rate for an investment opportunity?
Principles of Finance | Topic 5: Risk and Return | Slide 4/54

Outline

1 A first look at risk and return

2 Common measures of risk and return Risk and return

3 Historical returns of stocks and bonds

4 The historical trade-off between risk and return

5 Diversification in stock portfolios

Relation between them
6 Measuring systematic risk

7 Beta and the cost of capital

8 Portfolios of assets Portfolio risk and return

Principles of Finance | Topic 5: Risk and Return | Slide 5/54
A first look at risk and return

Outline

1 A first look at risk and return

2 Common measures of risk and return

3 Historical returns of stocks and bonds

4 The historical trade-off between risk and return

5 Diversification in stock portfolios

6 Measuring systematic risk

7 Beta and the cost of capital

8 Portfolios of assets
Principles of Finance | Topic 5: Risk and Return | Slide 6/54
A first look at risk and return

Observations:
• Small stocks outperform (see returns).
Figure: Risk and return • It has the largest fluctuations.
(Value of $100 invested at the end of 1925)
Explanation:
• Investors are averse to fluctuations.
• Demand a risk premium to bear risks.
Principles of Finance | Topic 5: Risk and Return | Slide 7/54
Common measures of risk and return

Outline

1 A first look at risk and return

2 Common measures of risk and return

3 Historical returns of stocks and bonds

4 The historical trade-off between risk and return

5 Diversification in stock portfolios

6 Measuring systematic risk

7 Beta and the cost of capital

8 Portfolios of assets
Principles of Finance | Topic 5: Risk and Return | Slide 8/54
Common measures of risk and return

The future returns of risk-free assets are certain. If your bank pays 4%
p.a., you know with certainty that your annual returns will be 4% in the
future.

The future returns of stocks are uncertain. The annual returns of stocks
will vary from one year to another. It may be 12% one year, -3% another,
etc.

Therefore, we can say the stock returns have probability distributions.

There are two very useful statistics regarding probability distributions:

1 Expected value: This is a measure for centrality. Location
2 Standard deviation: This is a measure for dispersion. Spread
 Spread

Location
Principles of Finance | Topic 5: Risk and Return | Slide 10/54
Common measures of risk and return

Expected return of asset i is:

N
µi = ∑ pj Rij
j =1

Variance of asset i is:

σi2= ∑ pj (Rij − µi)2

j =1

Standard deviation σi is the square root of variance.

In general, we cannot observe the true values of an asset’s expected

return, variance, and standard deviation.
Principles of Finance | Topic 4: Option Prices | Slide 36/39
The Black-Scholes option pricing model

Probability Rj

× R1
p1
p2 × R2

Asset i p3 × R3 N

∑ µi = ∑ pj Rij
. j =1
pN .
× .

RN
Principles of Finance | Topic 5: Risk and Return | Slide 12/54
Common measures of risk and return

Example 1

Suppose BFI stock currently trades for $100 per share. You believe that in
one year there is a 25% chance the share price will be $140, a 50% chance it
will be $110, and a 25% chance it will be $80. If BFI does not pay any
dividends, what is its expected return and what is the standard deviation of its
returns?

First, let’s calculate the returns:

1 If the price goes up to $140, the return is: Ru = ($140 − $100)/$100 = 40%.

2 If it goes up to $110, the return is: Rm = ($110 − $100)/$100 = 10%.

3 If it goes down to $80, the return is: Rd = ($80 − $100)/$100 = −20%.

Principles of Finance | Topic 5: Risk and Return | Slide 13/54
Common measures of risk and return

N N
µi = ∑ pj Rij σi2= ∑ pj(Rij − µi)2
j =1 j =1

Then, the expected return for BFI is:

µBFI = 0.25 × (40%) + 0.50 × (10%) + 0.25 × (−20%) = 10%

Finally, the standard deviation of returns for BFI is:

σBFI = ( 0.25 × (40% − 10%)2 + 0.50 × (10% − 10%)2 + 0.25 × (−20% − 10%)2 )0.5 = 21.2%
Principles of Finance | Topic 5: Risk and Return | Slide 14/54
Historical returns of stocks and bonds

Outline

1 A first look at risk and return

2 Common measures of risk and return

3 Historical returns of stocks and bonds

4 The historical trade-off between risk and return

5 Diversification in stock portfolios

6 Measuring systematic risk

7 Beta and the cost of capital

8 Portfolios of assets
Principles of Finance | Topic 5: Risk and Return | Slide 15/54
Historical returns of stocks and bonds

Figure 10.5: Empirical distribution of realised annual returns, 1926-2008

Principles of Finance | Topic 5: Risk and Return | Slide 16/54
Historical returns of stocks and bonds

The average return for an asset i is:

N
1
Ri = ∑ Rij (j represents period (e.g. year))
j =1 N

We can use Ri as an estimate of µi .

Average returns for the investments in Figure 10.5 are:

Principles of Finance | Topic 5: Risk and Return | Slide 17/54
Historical returns of stocks and bonds

The variance of returns for an asset i is:

N
∑ (Rij −Ri)2
si =
2 j=1

N−1

The volatility si is the square root of variance.

We can use si2and si as estimates of σi2and σi respectively.

Volatility of returns for the investments in Figure 10.5 are:

Principles of Finance | Topic 5: Risk and Return | Slide 18/54
Historical returns of stocks and bonds

Estimation error: using past returns to predict the future

Average return Ri can be obtained by computing returns from past data.

Ri is a good estimate of the future expected return µR, if the distributions

of past and future returns are the same.

Still, being an estimate, it is subject to estimation error, which is

measured by its standard error:
si
sei =
N

Typically, a 95% confidence interval for the expected return is:

Ri ± 2 × sei
Principles of Finance | Topic 5: Risk and Return | Slide 19/54
Historical returns of stocks and bonds

Example 2

Suppose that you are interested in knowing the expected return on S&P 500.
You look at historical returns during the past 83 years and estimate that
average histroical return is 11.6% with a standard deviation of 20.6%. What
is, then, the 95% confidence interval for the expected return on S&P 500?

This confidence interval is:

20.6%
11.6% ± 2 ( ) = 11.6% ± 4.5% (7.1%, 16.1%)
83

We are 95% confident that the expected return is between 7.1% and
16.1%.

Even with 83 years of data, we cannot estimate the expected return of

the S&P 500 very accurately. Limitations:
• Individual stocks tend to be volatile than large portfolios.
• Limited data: existed for only a few years.
Principles of Finance | Topic 5: Risk and Return | Slide 20/54
The historical trade-off between risk and return

Outline

1 A first look at risk and return

2 Common measures of risk and return

3 Historical returns of stocks and bonds

4 The historical trade-off between risk and return

5 Diversification in stock portfolios

6 Measuring systematic risk

7 Beta and the cost of capital

8 Portfolios of assets
Principles of Finance | Topic 5: Risk and Return | Slide 21/54
The historical trade-off between risk and return

Figure 10.6: The returns of large portfolios

Principles of Finance | Topic 5: Risk and Return | Slide 22/54
The historical trade-off between risk and return

Remember that our goal is to be able to calculate risk-adjusted expected

return, which is equal to rf plus a risk premium.

Figure 10.6 seems to suggest that the risk-adjusted expected return µi

for an asset i increases in proportion to its volatility σi .

Unfortunately, we observe that this simple suggestion does not hold

when we look at individual stocks.
Principles of Finance | Topic 5: Risk and Return | Slide 23/54
The historical trade-off between risk and return

Figure 10.7: The returns of individual stocks

(1926-2004)
Observations:
1.A relationship between size and risk
2.Even the largest stocks are more volatile than S&P 500
3.All individual stocks seem to have higher volatility and lower return
Large than we would have predicted from simple extrapolation from portfolios

Small
Principles of Finance | Topic 5: Risk and Return | Slide 24/54
The historical trade-off between risk and return

It is clear from Figure 10.7, that expected returns of individual stocks do

not increase in proportion to their volatilities.

Therefore, while volatility seems to explain expected returns of large

portfolios well, it fails to explain expected returns of individual stocks.

Why are portfolios less risky than individual stocks?

Principles of Finance | Topic 5: Risk and Return | Slide 25/54
Diversification in stock portfolios

Outline

1 A first look at risk and return

2 Common measures of risk and return

3 Historical returns of stocks and bonds

4 The historical trade-off between risk and return

5 Diversification in stock portfolios

6 Measuring systematic risk

7 Beta and the cost of capital

8 Portfolios of assets
Principles of Finance | Topic 5: Risk and Return | Slide 26/54
Diversification in stock portfolios

Firm-specific versus systematic risk

Usually, stock prices and dividends fluctuate due to either firm-specific

news or market-wide news.

Fluctuations due to firm-specific news is idiosyncratic risk.

Fluctuations due to market-wide news is systematic risk.

If you combine stocks into a large portfolio, you can diversify

idiosyncratic risk, but cannot get rid of systematic risk.
Principles of Finance | Topic 5: Risk and Return | Slide 27/54
Diversification in stock portfolios

Example 3

Suppose that there are two types of firms in an economy. Each year, the
economy has a 50-50 chance of being either strong or weak.
Type S firms are affected only by systematic risk. They earn a return of
40% if the economy is strong, and -20% if the economy is weak.
Type I firms are affected only by idiosyncratic risk, such that their returns
are equally likely to be 35% or -25%.
What are the expected returns and standard deviations of returns for an
S-type and an I-type firm?

Furthermore, suppose that you construct two portfolios. The first one PS
contains 10 S-type firms, and the second one PI contains 10 I-type firms.
What are the expected returns and standard deviation of returns for these
portfolios? (Note: Portfolio return and standard deviation are defined later in
the lecture.)
Principles of Finance | Topic 5: Risk and Return | Slide 28/54
Diversification in stock portfolios

The expected return for an S-type firm is:

N
µi = ∑ pj Rij µS = 0.5 × (40%) + 0.5 × (−20%) = 10%
j =1

The standard deviation of returns for an S-type firm is:

N
σi2= ∑ pj(Rij − µi)2 σS = ( 0.5 × (40% − 10%)2 + 0.5 × (−20% − 10%)2 )0.5 = 30%
j =1

The expected return and standard deviation of returns for PS are:

µPS = 10%
σPS = 30%

Note that σPS = σS . That is, there is no reduction in risk via

diversification.

σPS would not fall at all, if I continue to add more S-type firms into the
portfolio PS.
Principles of Finance | Topic 5: Risk and Return | Slide 29/54
Diversification in stock portfolios

The expected return for an I-type firm is:

µI = 0.5 × (35%) + 0.5 × (−25%) = 5%

The standard deviation of returns for an I-type firm is:

σI = ( 0.5 × (35% − 5%)2 + 0.5 × (−25% − 5%)2 )0.5 = 30%

The expected return and standard deviation of returns for PI are:

µPI = 5%
si 30%
σPI = sd (average return) = = = 9.5%
N 10

Note that σPI < σI . That is, there is some reduction in risk via
diversification.

σPI would fall all the way down to zero, if I continue to add more I-type
firms into the portfolio PI.
Principles of Finance | Topic 5: Risk and Return | Slide 30/54
Diversification in stock portfolios
Principles of Finance | Topic 5: Risk and Return | Slide 31/54
Diversification in stock portfolios

No arbitrage and the risk premium

Consider the I-type firms in Example 3. Because each individual I-type

firm is risky (i.e., σI = 30% > 0), should investors expect to earn a risk
premium when investing in this type of firms?

In a competitive market, the answer is NO. To see why, assume that the
risk-free rate was 3%. Then, I can borrow from a bank at 3% to construct
a very large portfolio of I-type firms.

This portfolio would have an expected return of µ PI = 5% and would

have almost no risk due to diversification.

This would be an arbitrage opportunity, which cannot exist in competitive

markets. Many investors would try to borrow from the bank, which would
push the risk-free rate up to 5%.
Principles of Finance | Topic 5: Risk and Return | Slide 32/54
Diversification in stock portfolios

This no-arbitrage argument leads us to the following principles:

1 The risk premium for diversifiable risk is zero.

2 The risk premium solely depends on a security’s systematic risk.

These principles imply that a stock’s volatility σ i , which is a measure of

total risk (= systematic risk + idiosyncratic risk), is not useful
determining the risk premium on individual securities.
Principles of Finance | Topic 5: Risk and Return | Slide 33/54
Diversification in stock portfolios

Consider again the I-type and the S-type firms in Example 3. We found
that they have the same volatility ( σI = σS = 30%), but different expected
returns.

S-type firm has a higher expected return µS = 10% > 5% = µI.

Then, the risk premium for the S-type firm is µS − rf = 10% − 5% = 5%.

And, the risk premium for the I-type firm is µI − rf = 5% − 5% = 0%.

How do we measure systematic risk?

Principles of Finance | Topic 5: Risk and Return | Slide 34/54
Measuring systematic risk

Outline

1 A first look at risk and return

2 Common measures of risk and return

3 Historical returns of stocks and bonds

4 The historical trade-off between risk and return

5 Diversification in stock portfolios

6 Measuring systematic risk

7 Beta and the cost of capital

8 Portfolios of assets
Principles of Finance | Topic 5: Risk and Return | Slide 35/54
Measuring systematic risk

In theory, market portfolio is the portfolio of all securities traded in the

capital markets.

In practice, a broad index, such as S&P 500 is used as a proxy for the
market portfolio.

Market portfolio is a well-diversified portfolio, which would contain only

systematic risk. All idiosyncratic risks are diversified away.

Then, we can measure a security’s systematic risk by determining the

sensitivity of its returns to the returns on the market portfolio.

This sensitivity is captured by the stock’s Beta β.

Principles of Finance | Topic 5: Risk and Return | Slide 36/54
Measuring systematic risk

The Beta of the market portfolio is equal to 1: β m = 1.

Stocks in cyclical industries such as high-technology tend to have Betas

higher than 1. For instance βApple = 1.89.

Stocks in non-cyclical industries such as retail tend to have Betas lower

than 1. For instance βWal−Mart = 0.31.

See Table 10.6 in the textbook for more data.

Principles of Finance | Topic 5: Risk and Return | Slide 37/54
Beta and the cost of capital

Outline

1 A first look at risk and return

2 Common measures of risk and return

3 Historical returns of stocks and bonds

4 The historical trade-off between risk and return

5 Diversification in stock portfolios

6 Measuring systematic risk

7 Beta and the cost of capital

8 Portfolios of assets
Principles of Finance | Topic 5: Risk and Return | Slide 38/54
Beta and the cost of capital

Capital asset pricing model (CAPM)

The difference between the expected return on the market portfolio rm

and the risk-free return rf yields the market risk premium:

Market risk premium = rm − rf

On the other hand, the risk premium for a single asset e is the difference
between the expected return on the asset r e and the risk-free return rf :

Risk premium for asset e = re − rf

CAPM postulates that asset e’s βe is the link between its risk premium
and the market risk premium:

re − rf = βe (rm − rf )
Principles of Finance | Topic 5: Risk and Return | Slide 39/54
Beta and the cost of capital

The expected return on an asset e is:

re = rf + βe(rm − rf )
Principles of Finance | Topic 5: Risk and Return | Slide 40/54
Beta and the cost of capital

Example 10.8 - 10.9

Suppose that the market portfolio tends to increase by 47% when the
economy is strong and decline by 25% when the economy is weak. The
economy is equally likely to be strong or weak. The risk-free interest rate is
5%. What is the expected return on an S-type firm, if βS = 0.833?
First, calculate the expected return on the market portfolio:

rm = 0.5 × (47%) + 0.5 × (−25%) = 11%

Now, calculate the expected return on S:

rS = rf + βS(rm − rf ) = 5% + 0.833(11% − 5%) = 10%

This is consistent with the figure we calculated in Example 3.

Principles of Finance | Topic 5: Risk and Return | Slide 41/54
Portfolios of assets

Outline

1 A first look at risk and return

2 Common measures of risk and return

3 Historical returns of stocks and bonds

4 The historical trade-off between risk and return

5 Diversification in stock portfolios

6 Measuring systematic risk

7 Beta and the cost of capital

8 Portfolios of assets
Principles of Finance | Topic 5: Risk and Return | Slide 42/54
Portfolios of assets

Portfolio return is a weighted average of individual asset returns:

n
RP = x1R1 + x2R2 + · · · + xnRn = ∑ xi Ri
i =1

Portfolio variance depends on covariances between pairs of individual

assets:
n n
σ2P = ∑ ∑ xi xj σij
i=1 j=1

How much risk could be eliminated in a portfolio depends on

the degree to which the stocks face common risks and their prices move together.
Principles of Finance | Topic 5: Risk and Return | Slide 43/54
Portfolios of assets

Example 4

Suppose that you have $1,000 to invest. You are considering to split your
money between Google and Microsoft. Google has an expected annual
return of 10% and volatility of 4%. Microsoft has an expected annual return of
6% and volatility of 2.5%. The covariance between these two stocks is -0.001.

What is the expected annual return and volatility of your portfolio, if you
invest (a) all of your money in Google, (b)$750 in Google, (c) half of your
money in Google, (d) 25% of your money in Google, (e) all of your money in
Microsoft? R =10% R =6% 1 2

σ1=4% σ2=2.5% σ12 = -0.001

Weights: X1 x2
(a) 1 0
(b) 0.75 0.25
(c) 0.5 0.5
(d) 0.25 0.75
(e) 0 1
Principles of Finance | Topic 5: Risk and Return | Slide 44/54
Portfolios of assets

n n n
RP = ∑ xi Ri σ𝟐𝐏 = ∑ ∑ xi xj σij
i =1 i =1 j=1

For part (a), if you invest all of your money in Google:

x1 R1 + x2 R2
µP = (1)(10%) + (0)(6%) = 10%
2
( X1 σ12 + X22 σ22 + 2 X1 X2 σ12 ) 0.5
σP = ((1)2(4%)2 + (0)2(2.5%)2 + (2)(1)(0)(−0.001) )0.5 = 4%

For part (b), if you invest $750 in Google:

µP = (0.75)(10%) + (0.25)(6%) = 9%

σP = ((0.75)2(4%)2 + (0.25)2(2.5%)2 + (2)(0.75)(0.25)(−0.001))0.5 = 2.38%

Principles of Finance | Topic 5: Risk and Return | Slide 45/54
Portfolios of assets

For part (c), if you invest half of your money in Google:

µP = (0.5)(10%) + (0.5)(6%) = 8%

σP = ((0.5)2(4%)2 + (0.5)2(2.5%)2 + (2)(0.5)(0.5)(−0.001) ) = 0.75%

0.5

For part (d), if you invest 25% of your money in Google:

µP = (0.25)(10%) + (0.75)(6%) = 7%

σP = ((0.25)2(4%)2 + (0.75)2(2.5%)2 + (2)(0.25)(0.75)(−0.001) )0.5 = 0.88%

Principles of Finance | Topic 5: Risk and Return | Slide 46/54
Portfolios of assets

For part (e), if you invest all of your money in Microsoft:

µP = (0)(10%) + (1)(6%) = 6%

σP = ((0)2(4%)2 + (1)2(2.5%)2 + (2)(0)(1)(−0.001) )0.5 = 2.5%

Note that as you invest more money in Microsoft, portfolio return linearly
decreases from 10% to 6%.

On the other hand, portfolio volatility initially decreases from 4%, but
then it starts increasing and reaches 2.5%.
 Efficient Frontier

10%

9%
A portfolio is inefficient whenever
8% it is possible to find another portfolio
that is better in terms of both expected
return and volatility.
7%

x
6%

5%
0% 1% 2% 3% 4%
Principles of Finance | Topic 5: Risk and Return | Slide 48/54
Portfolios of assets

CAPM states that the expected return on asset i should be:

Ri = Rf + βi (Rm − Rf )

Then, we can rewrite portfolio return Rp as:

n
RP = ∑ xiRi
i =1
n
= ∑ xi [Rf + βi (Rm − Rf)]
i =1
βP
n
= Rf + ∑ xi βi (Rm − Rf)
i =1
n
If we define the Beta of a portfolio as: βP = ∑ xi βi , then we have:
i=1

RP = Rf + βP(Rm − Rf )

Thus, CAPM can explain expected returns on portfolios as well as those

on individual assets.
Principles of Finance | Topic 5: Risk and Return | Slide 49/54
Portfolios of assets

CAPM also states that the total variance on stock i should be:

σi2 = βi2σm2 + σe2

2
βP

Then, we can rewrite portfolio variance as:

n n
2
σP2 = ∑ xi2 σi2 = ∑ xi (βi2σm2 + σe2)
i =1 i =1
n
σP2 = βP2σm2 + ∑ xi2 σe2
i =1

Thus, CAPM decomposes total risk of a portfolio into systematic risk and
idiosyncratic risk. The latter type of risk vanishes as the portfolio gets
larger.
Systematic risk
Idiosyncratic risk
Principles of Finance | Topic 5: Risk and Return | Slide 50/54
Portfolios of assets

Summary

CAPM is a model that is used to obtain expected returns of assets and

portfolios of assets.

The expected return of a risky asset contains a risk premium, since it is

assumed that the investors are risk averse.

The risk premium of an asset is determined by its systematic risk, and

not by its idiosyncratic risk.

Idiosyncratic risk is diversified without any cost in large portfolios, hence

this type of risk is not priced, and does not entail a premium.