Lecture 7:

Modern Portfolio Theory

Panagiotis Panagiotou

FINA1082 Principles of Finance

Department of Accounting and Finance
Greenwich Business School

November 9, 2021
[email protected]

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 1 / 63

Learning Objectives

By the end of this session, you will be able to:

Measure risk and return of individual assets.

Measure portfolio risk.

Understand the concept of risk diversification using portfolio investment.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 2 / 63

Measuring simple returns to holding a single share
Two sources of returns: dividends and capital gains

Calculating simple (percentage) returns: buy one share at t and pay Pt , sell the
stock at t + 1 and receive Pt+1 and Dt+1 .

Pt+1 − Pt + Dt+1 Pt+1 − Pt Dt+1

rt+1 = = +
Pt Pt Pt

Pt+1 −Pt
Pt
is referred to as the capital gain.
Dt+1
Pt
is referred to as the dividend yield.

The time between Pt and Pt+1 is called the holding period. The holding period
can be any amount of time: one second, two hours, three days, one week etc.

Often we write rt+1 = 100 × rt+1 , as 100rt+1 is the percentage of the gain with
respect to the initial capital Pt .

Since asset prices must always be non-negative (a long position in an asset is a

limited liability investment), the smallest value for rt+1 is -1 or -100%.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 3 / 63

Simple (percentage) Return: an example

If ABC plc is selling for £100 per share today and is expected to sell for
£110 one year from now, what is the return if the dividend one year
from now is forecasted to be £4.00?

Answer:

Pt+1 −Pt +Dt+1 110−100+4

rt+1 = Pt = 100 = 0.14 or 14%.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 4 / 63

Expected Return
In an uncertain world future cash flows are not known with certainty in
advance.

To calculate the expected return to holding a single asset, compute the

weighted average of all future possible returns, where the weight of each
possible future return is the respective probability of occurrence of each
future return.

S
X
Expected Return E(rt ) = P(s) × r (s)
s=1

P(s) = is the probability of occurrence of each of the s possible future states.

r(s) = is the return associated with each of the s possible future states.
S= is the number of all possible future states.

Disadvantage: probabilities can be subjective or difficult to be accurately

estimated.
Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 5 / 63
Expected Return: an example

Suppose you have predicted the following returns for stock X in four possible
states of the economy. What is the expected return?

Possible State of Economy Probability Return

Severe Recession 5% -52%
Mild Recession 25% -6.75%
Slow Growth 45% 14%
Moderate Growth 25% 31%

4
X
E(rX ) = P(s)r (s) ⇒
s=1
E(rX ) = (0.05)(−52%) + (0.25)(−6.75%) + (0.45)(14%) + (0.25)(31%) ⇒
E(rX ) = 9.76%

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 6 / 63

Question

Assume we have two assets: asset A and asset B.

Further assume that: E(rA ) = 12.00% and E(rB ) = 10.00%

Why would anyone want to invest in B when A is expected to have a

higher return?

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 7 / 63

Question

Assume we have two assets: asset A and asset B.

Further assume that: E(rA ) = 12.00% and E(rB ) = 10.00%

Why would anyone want to invest in B when A is expected to have a

higher return?

The answer depends on risk.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 7 / 63

Asset Risk

In finance, risk refers to the degree of uncertainty and/or potential

financial loss inherent in an investment decision.

Companies face risk from variability in project cash flows.

Investors face risk from variability in capital gains and dividends.

Rational aim is to minimise risk for given level of return.

To control risk it must be understood and measured.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 8 / 63

Asset Risk (cont.)
We measure the risk of an asset with its variance or, alternatively,
standard deviation (the standard deviation is just the square root of the
variance).
PS
Variance (Var): σ 2 = s=1 P(s) × (r (s) − E(r ))2 , E(r) is the mean
return.
√
Standard Deviation (Std): σ = σ2

Variance measures the probability weighted average of an investment’s

outcomes around the mean.

As such, it measures dispersion of outcomes around the average.

Higher variance, high dispersion around the average, more risk.

Note: variance is not the only possible way to measure risk, but it is a popular
choice in finance. A nice feature of standard deviations is that they are in the
same units as the original variable.
Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 9 / 63
Asset Risk: an example
Suppose you have predicted the following returns for stock X in four possible
states of the economy. What is the asset risk? Recall that E(rX ) = 9.76%.

Possible State of Economy Probability Return

Severe Recession 5% -52%
Mild Recession 25% -6.75%
Slow Growth 45% 14%
Moderate Growth 25% 31%

4
X
σX2 = P(s) × (r (s) − E(r ))2 ⇒
s=1

σX2 = (0.05)(−0.52 − 0.0976)2 + (0.25)(−0.0675 − 0.0976)2 +

(0.45)(0.14 − 0.0976)2 + (0.25)(0.31 − 0.0976)2 ⇒
σX2 = 0.038
q √
and σX = σX2 = 0.038 = 0.195 or 19.5%

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 10 / 63

Asset Risk (cont.)

Often we are not given probabilistic information on an investment and its

outcomes, but we are given historical data on its performance. In this
situation, we calculate a sample variance to measure risk.

Sample Variance definition: you are given historical, annual (for

example) returns on an investment over K years. Call the return in the
first year R1 , in the second year R2 , ... and the return in the Kth year is
RK . The sample variance is:

K
1 X
s2 = (Ri − R̃)2
K −1
i=1

where R̃ is the sample average and the sample standard deviation is just
the square root of the sample variance.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 11 / 63

Sample Variance: an example

Suppose that you are told that in the most recent 4 years, the nominal annual
returns on the UK equity market have been 13.70%, 35.80%, 45.14%, and
-8.88% respectively. What is the sample variance?

Answer:

The sample mean return over this period is 21.40%, so:

1
s2 = × ((0.1370 − 0.2144)2 + (0.3580 − 0.2144)2
3
+ (0.4514 − 0.2144)2 + (−0.0888 − 0.2144)2 ) = 0.0582

√ √
Thus, s = s2 = 0.0582 = 0.2412 or 24.12%.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 12 / 63

Covariance

Variance and standard deviation measure the variability of individual

assets.

We also interested to measure the relationship between the return on

one asset and the return on another.

Covariance is a statistical measure that intends to capture the degree to

which two random variables tend to move together.

The sign of the covariance shows the tendency in the linear relationship
between the variables.

Cov (X , Y ) = σX ,Y = E[(X − µX )(Y − µY )]

where µX and µY are the means of the random variables X and Y.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 13 / 63

Correlation Coefficient (ρ)

The actual value of the covariance is not very meaningful because its
measurement is extremely sensitive to the scale of the two variables.

Also, the covariance may range from negative to positive infinity.

For these reasons, we calculate the correlation coefficient, which

converts the covariance into something that is more useful and intuitively
appealing.

Cov (X , Y )
ρX ,Y = Corr (X , Y ) = p
Var (X )Var (Y )
Alternatively,

p
Cov (X , Y ) = ρX ,Y × Var (X )Var (Y ) = ρX ,Y × σX σY

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 14 / 63

Correlation Coefficient (cont.)

The correlation coefficient is a measure of linear association between

two variables and lies between -1 and +1.

It shows the direction and strength between the variables and, in fact, it
is the normalized version of the covariance:

- ρ = −1 indicates a perfect negative association

- ρ = +1 indicates a perfect positive association
- ρ = 0 indicates no linear association between the two variables

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 15 / 63

Correlation Coefficient (cont.)

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 16 / 63

Rationality

Other things equal, rational investors are assumed to:

prefer more return to less.

prefer money now to later.

prefer to avoid risk.

they make decisions that promote their self-interest.

always make completely rational decisions.

be able to make rational decisions among thousands of alternatives.

But how investors rank various combinations of risky alternatives?

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 17 / 63

Preferences: Which investment would you chose?

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 18 / 63

The concept of risk aversion

Assume an individual is offered the following two alternatives:

1 Receive £50 with certainty.

2 Take part in a gamble, with 50% chance of winning £100 and a 50%
chance of getting nothing.

The expected value of both alternatives is £50.

What will an investor choose? There are three possibilities:

- The investor chooses £50 with certainty.

- The investor chooses the gamble.
- The investor is indifferent.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 19 / 63

Risk Averse Investor

If the investor chooses £50 with certainty, she/he is said to be risk

averse.

The investor is risk averse because she/he does not want to take the
chance of getting anything at all.

In general, investors are likely to shy away from risky investments for a
lower, but guaranteed return.

That is why risk averse investors wants to minimize their risk for the
same amount of return, and maximize their return for the same amount
of risk.

In many finance models and applications, the representative investor is

assumed to be risk-averse.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 20 / 63

Risk Loving Investor

If the investor chooses the gamble, she/he is said to be risk loving or

risk seeking.

The gamble has an uncertain outcome, but with the same expected
value as the guaranteed outcome.

Thus, an investor choosing the gamble means that the investor gets
extra “utility” from the uncertainty associated with the gamble.

Would the investor be willing to accept a smaller expected value

because he gets extra utility from risk?

Indeed, risk seekers will accept less return because of the risk that
accompanies the gamble.

Gambling has a negative expected value.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 21 / 63

Risk Neutral Investor

If an investor is indifferent about the gamble or the guaranteed outcome,

then the investor may be risk neutral.

Risk neutrality means that the investor cares only about return and not
about risk, so higher return investments are more desirable even if they
come with higher risk.

Many investors may exhibit characteristics of risk neutrality when the

investment at stake is an insignificant part of their wealth.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 22 / 63

Indifference Curves

A risk-averse investor would rank the guaranteed outcome of £50 higher

than the uncertain outcome with an expected value of £50.

We can therefore say that the utility (or satisfaction or happiness) that an
investor derives from the guaranteed outcome of £50 is greater than the
utility (s)he derives from the alternative

More generally, investors derive satisfaction (or utility) from their

preferences for risk and return.

Utility is a measure of relative satisfaction that an investor derives from

investing in an asset or a portfolio.

An indifference curve plots the combinations of risk and return pairs

that an investor would accept to maintain a given level of utility.

All securities that lie on the same indifference curve, give an investor
equal satisfaction thereby making the investor indifferent.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 23 / 63

Indifference Curves (cont.)

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 24 / 63

Indifference Curves (cont.)
Combinations of risk and return a and c on indifference curve C2 have
the same level of utility for the investor.
Thus, the investor is indifferent between those two combinations, and
does not care whether (s)he is at point a or point c on indifference curve
c2.
However, indifference curves C1 and C3 represent different levels of
utility for the investor.
A rational investor, (s)he would prefer combination d to c, as for the same
level of risk, it offers a higher expected return and thus a higher utility.
Thus, indifference curve C3 refers to a higher utility level and
indifference curve C1 refers to a lower utility level as compared to
indifference curve C2.
The utility of a risk-averse investor always increases as you move
northwest (higher return with lower risk). Because all investors prefer
more utility to less, investors want to move to the indifference curve with
the highest utility.
Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 25 / 63
Indifference Curves (cont.)
Upward sloping at an increase rate: in order to take on progressively
more risk, progressively higher returns are required to keep their utility
constant.

The greater the slope, the higher is the risk aversion of the investor as a
greater increment in return is required to accept a given increase in risk.

Thus, the slope coefficient of an indifference curve is closely related to

the risk aversion coefficient.

The most risk-averse investor has an indifference curve with the greatest
slope.

An investor that is less risk averse has a much flatter indifference curve
as their demand for increased returns as risk increases is much less
acute.

The risk-loving investor’s indifference curve, however, exhibits a negative

slope, implying that the risk-lover is happy to substitute risk for return.
Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 26 / 63
Indifference curves for investors with different levels of
risk aversion

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 27 / 63

Risk and Return: Before Harry Markowitz

Before Harry Markowitz (JF, 1952) introduced modern portfolio theory

(MPT) and Eugene Fama and others proposed the theory of efficient
markets, investment strategies were based on the skills in picking
winners.

If the focus is to pick winners, risk is measured with the standard

deviation or with variance.

Assume we have the following information regarding gold and FTSE100:

Asset Return Volatility (std)

Gold 8.8% 20.8%
FTSE100 12.8% 18.3%

Which asset would you invest in?

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 28 / 63

Risk and Return: Why Buy Gold?
Assume we have the following information regarding gold and FTSE100:

Asset Return Volatility (std)

Gold 8.8% 20.8%
FTSE100 12.8% 18.3%

Gold has a lower return and a higher volatility than the FTSE100.

Can we say that buying gold is a bad investment?

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 29 / 63

Risk and Return: Why Buy Gold?
Assume we have the following information regarding gold and FTSE100:

Asset Return Volatility (std)

Gold 8.8% 20.8%
FTSE100 12.8% 18.3%

Gold has a lower return and a higher volatility than the FTSE100.

Can we say that buying gold is a bad investment?

The correlation coefficient (ρ) between FTSE100 and gold is -0.40.

Why does that matter? It matters because we can choose to invest in

combinations of different assets and not just on individual assets.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 29 / 63

Combining Assets in Portfolios
A portfolio is a specific combination of assets, such as stocks and
bonds, that are held by an investor.

One convenient way to describe a portfolio is by listing the proportion of

the total value of the portfolio that is invested into each asset.

These proportions are called portfolio weights (wi ):

amount invested in asset i

wi = total amount invested

Portfolio weights are usually expressed in percentages.

A portfolio is uniquely defined by the portfolio weights.

PN
Portfolio weights sum to one, i.e. i=1 wi = 1, where N is the total
number of assets in the portfolio.
Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 30 / 63
Buy a Portfolio with Gold

Alternative Investment: holding a portfolio of gold and shares.

Volatility looks like following a U-shaped pattern.

Volatility is always lower if we invest in both of these assets.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 31 / 63

Buy a Portfolio with Gold (cont.)

Conclusion: Although gold in isolation as an asset is very risky, when it is

held within a portfolio, gold reduces substantially the total portfolio risk,
in that it works as insurance against some forms of risk (inflation, foreign
exchange risk, ...).

This result is due to this negative correlation with the stock market
returns.

What it means is there are times when the stock market does very
poorly and those tend to be times when gold does well.

Investors value this feature as in bad times when the stock market
collapses it is quite useful to have into your portfolio an asset which
somehow mitigates that collapse, makes the losses smaller and reduces
your downside risk.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 32 / 63

How do we calculate portfolio returns?
Asset Return Volatility (std)
Gold 8.8% 20.8%
FTSE100 12.8% 18.3%

Assume we invest 40% of our capital in gold and the remaining 60% in
FTSE100:

Portfolio Return: RP = (wgold × Rgold ) + (wFTSE100 × RFTSE100 ) =

(0.4 × 8.8%) + (0.6 × 12.8%) = 11.2%

Essentially, the portfolio return it’s just a linear combination of returns

with a weight of 40% and 60%.

More generally, the portfolio return is simply the weighted sum of the
individual asset returns where the weights represent the percentage out
of the total capital invested in each asset.
Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 33 / 63
How do we calculate portfolio risk?

Asset Return Volatility (std)

Gold 8.8% 20.8%
FTSE100 12.8% 18.3%

If it was linear:

Portfolio Risk: σP = (wgold × σgold ) + (wFTSE100 × σFTSE100 ) =

(0.4 × 20.8%) + (0.6 × 18.3%) = 19.3%

But, volatility is non-linear!

σP2 2
= (wgold 2
σgold 2
) + (wFTSE100 2
σFTSE100 ) + 2(wgold σgold )ρ(wFTSE100 σFTSE100 ) =
2 2 2 2
(0.4 × 0.208 ) + (0.6 × 0.183 ) + 2 × 0.4 × 0.6 × (−0.40) × 0.208 × 0.183 =
0.0117 q √
Thus, σP = σP2 = 0.0117 = 0.108 or 10.8% (< 19.3%)

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 34 / 63

Mean and Variance of a two-asset portfolio

Assume two risky assets. We form a portfolio with w1 being the weight of
asset 1 and w1 + w2 = 1 ⇒ w2 = 1 − w1 being the weight of asset 2.

RP = w1 R1 + w2 R2

E(RP ) = E(w1 R1 + w2 R2 ) = E(w1 R1 ) + E(w2 R2 ) = w1 E(R1 ) + w2 E(R2 )

The formula of the variance of portfolio returns is:

Var(RP ) = w12 σ12 + w22 σ22 + 2w1 w2 ρ1,2 σ1 σ2

Where did this formula come from?

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 35 / 63

Mean and Variance of a two-asset portfolio (cont.)

Proof:

Var(RP ) = σP2 = E((RP − E(RP ))2 )

= E((w1 R1 + w2 R2 − E(w1 R1 + w2 R2 )2 )
= E((w1 R1 + w2 R2 − w1 E(R1 ) − w2 E(R2 ))2 )
= E((w1 R1 − w1 E(R1 ) + w2 R2 − w2 E(R2 ))2 )
= E((w1 (R1 − E(R1 ) + w2 (R2 − E(R2 )))2 )
= E(w12 (R1 − E(R1 ))2 + w22 (R2 − E(R2 ))2 +
+ 2w1 w2 (R1 − E(R1 )(R2 − E(R2 ))
= w12 E(R1 − E(R1 ))2 + w22 E(R2 − E(R2 ))2 +
+ 2w1 w2 E(R1 − E(R1 )E(R2 − E(R2 ))
= w12 σ12 + w22 σ22 + 2w1 w2 ρ1,2 σ1 σ2

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 36 / 63

Expected Return of an N-asset portfolio

Recall that a portfolio is uniquely defined by the portfolio weights.

Suppose there are N assets, i = 1, 2, 3, ... , N

Given the definition of portfolio weights wi as:

amount invested in asset i

wi = total amount invested

The portfolio expected return is equal to the weighted sum of the returns
on the individual assets in the portfolio.

N
X
E(RP ) = wi E(Ri )
i=1

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 37 / 63

Variance of an N-asset portfolio

The portfolio variance depends on both variances of the individual

assets in the portfolio and their covariances.

σP2 = Var (RP )

N
X N
X N
X
= wi2 Var (Ri ) + wi wj Cov (Ri , Rj )
i=1 i=1 j=1,i6=j

N
X N
X
= wi2 σi2 + wi wj σi,j
i=1 j=1,i6=j

The covariance terms capture how the co-movements of returns affect

portfolio variance.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 38 / 63

The Relative Importance of Variances and Covariance
What happens if I increase the number of assets in my portfolio?
1
Suppose an equally-weighted (i.e. wi = N ), well diversified portfolio (i.e.
N tends to infinity).
N
X N
X
σP2 = wi2 σi2 + wi wj σi,j
i=1 j=1,i6=j

N N N
X 1 X X 1
≈ ( )2 σi2 + ( )2 σi,j
N N
i=1 i=1 j=1,i6=j

1 2 1
=( ) NVar + ( )2 (N 2 − N)Cov
N N

1 1
=( )Var + (1 − )Cov ≈ Cov
N N
Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 39 / 63
The Relative Importance of Variances and Covariance

If I have a very large of assets in a portfolio (strictly speaking an infinite

number), and;

I invest exactly the same amount of money in all of these assets (i.e. not
taking a particularly strong stand on any asset);

Then the variance of the returns by any particular assets becomes

irrelevant;

The only thing that matters, is the covariance between the assets.

This is the key insight of the modern portfolio theory.

What really matters to find the risk of a portfolio is the covariance

between assets, it’s not the variance itself.

And by diversifying, what you get is that you eliminate any risk which is
specific to a particular stock.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 40 / 63

Diversification
Investors can reduce portfolio risk by holding combinations of assets
whore returns do not move exactly together, i.e. when they are less than
perfectly correlated.

Investors can reduce their exposure to individual asset risk by holding a

diversified portfolio of assets.

Diversification is the spreading of investable funds across different

assets so as to eliminate some risk.

“Don’t put all your eggs in one basket”.

Diversification may allow for the same portfolio expected return with
reduced risk.

To maximize the benefits of diversification, we should look for assets that

tend not to move together: the lower the correlation, the more
diversification we achieve.
Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 41 / 63
Diversification (cont.)
Total portfolio risk can be divided into systematic and unsystematic risk.

Market risk: is due to systematic factors such as changes in interest

rates, business cycles and government policy. Also known as
“systematic risk” or “undiversifiable risk”. This risk cannot be
eliminated through diversification. Examples:
- It is due to the economy-wide sources of risk that affect the overall market:
uncertainty, GDP growth, interest rates, inflation, exchange rates, market
liquidity, government legislation.
- other external factors over which the company has no control.

Specific risk: is specific to a given company or asset. Also known as

“unsystematic risk” or “diversifiable risk”. This risk is eliminated
through diversification. Examples:
- announcements specific to a company e.g. chief executive resigns.
- strikes due to bad labour relations.
- improved competitor products, management quality etc.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 42 / 63

Limits to Diversification

Total Portfolio Risk = Market Risk + Specific Risk

Given that the majority of financial assets are positively correlated,

diversification can reduce risk, but not completely eliminate risk.

Assuming a well-diversified portfolio, specific risk can be eliminated and

be equal to zero.

However, market risk cannot be eliminated through diversification.

Therefore for a well diversified investor only market risk matters!

The diversified investor would want to be compensated for the risk that
(s)he can not avoid by diversification!

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 43 / 63

Limits to Diversification: Specific Risk

As the number of securities in the portfolio increases, specific risk tends

to decrease.
This reflects the fact that, on average, securities are less than perfectly
correlated.
In the limit, as the number of securities becomes very large, specific risk
is eliminated and the remaining risk reflects market risk only.
Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 44 / 63
Limits to Diversification: Market Risk

Market risk cannot be eliminated through diversification.

Lowest bound of total portfolio risk = market risk (i.e., Cov).
Therefore for a well diversified investor only market risk matters!
Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 45 / 63
Many Risky Assets

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 46 / 63

Investment Opportunity Set

The set of assets, in which I am allowed to invest given the asset universe
available to me, is known in the literature as the envelope curve.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 47 / 63

Mean - Standard Deviation Trade-off

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 48 / 63

Mean - Standard Deviation Trade-off (cont.)

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 49 / 63

Efficient Frontier

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 50 / 63

Efficient Frontier (cont.)

The Minimum variance frontier is the set of portfolios that minimizes the
portfolio standard deviation for a given level of expected retutn.

The Efficient frontier is the set of portfolios that maximizes the return for
a given portfolio standard deviation.

The efficient frontier is the upward sloping part of the envelope curve,
starting from the minimum variance portfolio.

Rational investors will only invest in those portfolios on the efficient

frontier.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 51 / 63

Portfolios: One Risky Asset + One Risk-Free Asset

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 52 / 63

Portfolios: Many Risky Assets + One Risk-Free Asset

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 53 / 63

Portfolios: Many Risky Assets + One Risk-Free Asset
Point P is vastly inefficient: for the same level of risk, I can get a higher
return by moving up on straight line emanating from the risk-free rate of
return.

The highest possible slope is going to be the straight line which is

tangent to the efficient frontier.

Any point above (and to the left) of point T is not achievable, and point
below (and to the right) is inefficient.

The point of tangency T defines the tangency portfolio.

The tangency portfolio defines the set of risky assets, which can be
combined with the risk-free asset, and defines the best set of
opportunities in terms of risk-return trade-off that are achievable in a
world where you can invest in the risk-free asset.
Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 54 / 63
Two Fund Separation Theorem

Suppose investors have homogenous expectations.

Then all investors, regardless of taste, risk preferences, and initial wealth
will hold a combination of two assets: a risk-free asset and an efficient
(optimal) portfolio of risky assets (i.e. the tangency portfolio).

The separation theorem allows us to divide an investor’s investment

problem into two distinct steps: the investment decision and the
financing decision.

Investment decision: identify the optimal risky portfolio.

Financing decision: lend or borrow at the risk-free rate.

Therefore, the individual investor’s risk preference determines the

amount of financing:
- a risk averse investor would place a larger weight on the risk-free asset
(lending to the government instead of investing in the optimal risky portfolio)
- borrowing to purchase additional amounts of the optimal risky portfolio.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 55 / 63

Market Portfolio

The two-fund separation theorem implies that all investors will hold the
same portfolio of risky assets.

In equilibrium, supply must equal demand for all assets. Thus, in

equilibrium, the tangency portfolio must be the market portfolio (with
weights given by the market value of the asset relative to the total market
value of all markets).

Tangency portfolio = market portfolio (M) = wealth of the economy.

Earlier we said that risk must be measured with reference to a

benchmark portfolio.

If, in equilibrium, all investors are holding the market portfolio of risky
assets, it is natural to measure risk with reference to this portfolio.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 56 / 63

Capital Market Line (CML)

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 57 / 63

Capital Market Line (cont.)
The Capital Market Line (CML) is the tangent line drawn from the point
of the risk-free asset to the feasible region of risky assets.

The point of tangency (M) between the CML and the efficient frontier is
known as the Market Portfolio.

Market portfolio is the optimal combination of risky assets given the

existence of the risk-free asset.

Investors are able to move along the CML by changing the proportions
of the risk-free asset and the market portfolio.

Given the risk-free asset, the CML becomes the new efficient frontier
(CML portfolios dominate those on the efficient frontier).

The investor’s risk preferences will determine the choice of CML

portfolio.
Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 58 / 63
Portfolio Choice

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 59 / 63

Implementing Portfolio Optimization

In practice, implementing portfolio optimization as a practical tool is not that

straightforward:

It is a single period model;

Borrowing at the risk-free rate is an unrealistic assumption;

Identifying the market-portfolio;

Constructing the market-portfolio;

It ignores transactions costs;

It assumes normally distributed asset returns;

Portfolio composition is usually very sensitive to forecasts of expected

returns (GIGO: garbage in, garbage out).

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 60 / 63

Key Points

Diversification reduces total portfolio risk.

In well diversified portfolios, covariances among stocks are more

important than individual assets.

Risk is not about the standard deviation of the individual asset but it is
about covariance risk, that is the source of risk that matters. It is the
marginal contribution of an asset when you add it to your portfolio.

Investors should try to hold portfolios on the efficient frontier.

With a riskless asset, all investors should hold the tangency portfolio.

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 61 / 63

Before you go

Post any questions on QnA Forum or send email at:

[email protected]

What’s next? Capital Asset Pricing Model (CAPM)

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 62 / 63

The End

See you all next week!

Panagiotis Panagiotou (Greenwich) Modern Portfolio Theory FINA1082 (Term 1) 63 / 63