Module 2

Module 2

Decision-making under risk and

uncertainty
Introduction
Until now, we have examined managerial decision-making primarily
under conditions of certainty. In such cases, the manager knows exactly
the outcome of each possible course of action. However, in business the
outcome of a decision is usually far from certain (at the time the
decision is taken) because the decision maker has incomplete
information and the outcome depends on the simultaneous behaviour of
rival firms and other factors influencing the managerial cost and
demand conditions. When the outcome of a decision is not predictable
with certainty, we say that the decision is made under conditions of risk
or uncertainty. Most strategic decisions of the firm are of this type.
Therefore it is essential to extend the basic model of the firm presented in
Module 1 to include risk and uncertainty.

In the first module, we distinguished between risk and uncertainty and

introduced some of the concepts essential for risk analysis. Building on
that discussion, in this module, we examine methods for measuring risk
and for analysing the manager’s attitude toward risk.

Upon completion of this module you will be able to:

 explain the difference between risk and uncertainty.

 measure the expected return and the measure of risk.
 explain the concept of cardinal utility as it pertains to risk.
Outcomes  identify attitudes towards risk identified by the behaviour of the
marginal utility of income.
 determine the risk premium.
 explain how decision makers adjust for risk in their estimation of
projects’ rates of return.
 explain the concepts of asymmetric information as reflected by
adverse selection and moral hazard.

26
 E5 Managerial Economics

Asymmetric Situations in which one party knows more about

information: its own actions or personal characteristics than
another party. When some people in the market
have better information than others, the people
with the least information may choose not to
Terminology participate in a market.

Certainty: Exists if the outcome of a decision is known in

advance without a shadow of a doubt.

Marginal utility: The change in total utility that takes place when
one more unit of money is gained or lost.

Risk: When the probabilities of each outcome can be

assigned on an objective basis.

Uncertainty: The case when there is more than one possible

outcome to a decision and where the probability of
each specific outcome occurring is not known or
even meaningful.

Risk and uncertainty

In simple microeconomics, economists assume full information, or
certainty. That is, they assume they know the exact shape and location of
demand and cost curves, such that they know exactly how much will be
demanded at each price and exactly what the cost of production will be at
the chosen output level. In the business world, however, firms typically
operate under conditions of incomplete information, or uncertainty, and
must estimate the quantity demanded and the costs of production based
on the limited information they have at hand or can obtain by conducting
information-search activity.

The state of information under which a decision is made has important

implications for the predictability of the outcome of that decision. If there
is full information, the outcome of a decision will be foreseen clearly and
unambiguously. In this situation (of certainty), the firm can accurately
predict the outcome of each of its decisions. When there is less than full
information, however, the decision maker may foresee several potential
outcomes to a decision and, therefore, will be unable to predict
consistently which outcome will actually occur. In this case, we say that
the individual or firm is operating under conditions of risk and
uncertainty.

Certainty exists if the outcome of a decision is known in advance without

a shadow of a doubt. Therefore, in this case, there is only one possible
outcome to a decision and this outcome is known precisely. On other
hand, uncertainty is the case when there is more than one possible
outcome to a decision and where the probability of each specific outcome
occurring is not known or even meaningful.

27
Module 2

Risk can be regarded as a subcategory of uncertainty in which the

probabilities of each outcome can be assigned on an objective basis. Risk
is involved when one flips a coin, or throws dice. The probability of
flipping a coin and having it land ‘heads’ is 1/2, since there are only two
possible outcomes, and each is equally likely to occur, given an unbiased
coin. Similarly, when one throws two dice, the probability that they will
turn up ‘double six’ or any other pair of numbers, is 1/6 x 1/6 = 1/36.

Measuring risks with probability distributions

As the above examples suggest, the greater the variability – the greater
the number and range of possible outcomes – the greater is the risk
associated with the decision or action. In the example, the probability of
heads or tails is 1/2, whereas the probability of dice landing a pair is 1/36.
In more complex business decisions, such as drilling for oil, it is possible
that due to insufficient information or instability in the structure of the
relevant variables, the investor will not know either the possible oil
outputs or their probability of occurrence.

In the previous section, we defined risk as the situation where there is

more than one possible outcome to a decision and the probability of each
possible outcome is known or can be estimated. In this section we
examine the meaning and characteristics of probability distributions, and
then we use these concepts to develop a precise measure of risk.

Probability distributions
When faced with outcomes that involve risks, a primary task of managers
is to develop techniques that will enable them to calculate and
subsequently minimise the risks inherent in a particular problem. One
method used to accomplish this is to calculate the probability distribution
of possible outcomes from a set of sample observations, and then
compute an expected value; that is, if several different levels of profit (or
loss) are perceived as possible and each of these is assigned a probability
of occurring. How does the decision maker summarise all these data so
they can be compared with other potential solutions to the same problem?

The probability of an event is the chance or odds that the event will occur.
For example, suppose you have just made an investment of $100 value in
shares of a car-making (low technology) company believing that one of
the three following possibilities might occur to your investment:
1. Its value increases by 10 per cent, to $110.
2. Its value stays the same, ($100).
3. Its value falls by 10 per cent, to $90 (see Table 2-1).

Furthermore, assume that, based on your prior investigation of this

company’s past performance as well as the current market conditions, you
have come to conclude that the probability (likelihood) of outcome (a) is
25 per cent, that is a one out of four chance that this outcome occurs. For

28
 E5 Managerial Economics

outcome (b) and (c) these probabilities are 50 per cent and 25 per cent,
respectively.
Table 2-1 Probability distribution of states of alternative scenarios of
share prices (low-tech company)

State of the affairs Probability of

occurrence
Share price increases 0.25
Share price stays unchanged 0.50
Share price decreases 0.25
Total 1.00

Note that the sum of the probabilities is l, or 100 per cent, since one of the
three possible scenarios of the share prices must occur with certainty.

The probability distribution depicts all possible payoffs and their

associated probabilities. Figure 2-1, below, shows the probability
distribution of your car-making company’s stock price. Each bar
represents a different possible payoff from investing in the company’s
shares.
Figure 2-1

Expected value
Given the probabilities associated with the possible outcomes of your
risky investment, how much can you expect to make? The answer to this
question is the expected value. The expected value of a lottery is a
measure of the average payoff that the lottery will generate. The expected
value of an outcome is the value of that outcome multiplied by the
probability of that outcome occurring. Since several outcomes are
possible under risk and uncertainty, the expected value of a decision is the
sum of the expected values of all the possible outcomes that may follow
the decision. We can illustrate this with your car-maker stock example:

29
Module 2

E ( X )  P1 X 1  P2 X 2  ............Pn X n
n (1)
E ( X )   Pi X i
i 1

where Xi is the value of the ith payoff, and Pi is the probability of the ith
state of nature. Or
Expected value = Probability of (a) x Payoff if (a) occurs
+ Probability of (b) x Payoff if (b) occurs
+ Probability of (c) x Payoff if (c) occurs.

Applying this formula we get:

Expected value = (0.25 x 110) + (0.50 x 100) + (0.25 x 90) =100

Measures of risk (standard deviation)

Suppose you had a choice of two investments, Table 2-2 presents the
payoffs associated with two investments: L and H. Investment L is the
same as the investment in the shares of the (low tech) car-making
company in Table 2-1, whereas H represents the outcome of investment
in another company’s shares, e.g., a (high tech) telecommunication
company. The expected value of each project is assumed to be $100, but
the range of outcomes for project L (from $90 to $110) is smaller than for
project H. The latter is assumed to range from $50 to $150. These are
obtained from a greater share price increase or decrease, which is
assumed to be 50 per cent, instead of 10 per cent (as in project L). The 50
per cent increase or decrease in share prices translates into the value of
your investment increasing or decreasing from $100 to $150 or $50,
respectively. The probabilities by which the three outcomes are expected
to occur are kept the same in both projects.
Table 2-2 Probability distribution of states of share prices (high and
low tech companies)

Project Sate of share Probability Outcome Expected

prices of value
occurrence

Increase 0.25 $110 $27.50

L Unchanged 0.50 $100 $50.00

Decrease 0.25 $90 $22.50

Expected value from Project L $100.00

Increase 0.25 $150 $37.50

H Unchanged 0.50 $100 $50.00
Decrease 0.25 $50 $12.50
Expected value from Project H $100.00

30
 E5 Managerial Economics

Figure 2-2 depicts the expected value and the variability in the outcomes
of investment L, in panel (a) and H, in panel (b). Note that the expected
values of the two investments are the same: $100. However, the
telecommunication stock is riskier than the car-maker’s stock because
while the stock of car-maker will probably remain at its current value of
$100, the telecommunication stock has a greater likelihood of going up or
down, panel (b).
Figure 2-2

Again, the height of each bar measures the probability that a particular
outcome (measured along the horizontal axis) will occur. Since both
investments have the same expected outcome ($100) but the relationship
between the payoffs (outcomes) is less dispersed in investment L, panel
(a), than in investment H, panel (b), investment L is less risky than H. In
other words, with the telecommunication stock, the investor stands to gain
more or lose more than with stock in the car-making company.

Intuitively, we sense that the farther away from the mean the actual
payoff can be, the riskier the investment. Hence, one way of measuring
risk is to calculate the range, which is the difference between the most
extreme payoff values. In our example, as noted above, the range of
investment L is 20 (from a low 90 to a high 110) while the range of
investment H is 100 (from a low 50 to a high 150).

The range, however, is useful in preliminary evaluation, but it considers

only the extreme values and gives no weight to values in between. A
more common and more accurate measurement of risk is the statistic
called standard deviation, which is a measurement of variation of payoffs
from the expected value. The higher the standard deviation, the greater
the deviation of possible payoffs, and therefore, the greater the risk.
The standard deviation is calculated as follows:
  [ X i  E ( X i )]. P i (3)

where  is the standard deviation. This expression suggests a three-step

procedure in calculating the standard deviation: (a) first calculate the

31
Module 2

expected value, also called the mean, (b) then take the difference between
each outcome (payoff) and the mean, and then square the result. And (c)
finally multiply each squared deviation in step (b) by the associated
probability and add them up.

Demonstration problem
Find the standard deviation of the two investment alternatives L and H, in
the example above.
Answer:

Investment L: Expected value = $100

Payoff (Xi) Prob.(%) [Xi- E(X)]2 [Xi- E(X)] 2.Pi

90 0.25 [90 –100] 2= 100 100 x 0.25 = 25
2
100 0.50 [100 –100] = 0 0 x 0.50 = 0
2
110 0.25 [110 –100] =100 100 x 0.25 = 25

Standard Deviation = 50  7

Investment H: Expected value = $100

50 0.25 [50 –100] 2= 2500 2500 x 0.25 = 625

2
100 0.50 [100 –100] = 0 0 x 0.50 = 0
2
150 0.25 [150 –100] =2500 2500 x 0.25 = 625

Standard Deviation = 1250  35

Based on these calculations, investment H, with a greater

standard deviation, is far more risky than investment L.

An alternative measure of the riskiness of a risky investment is

the variance. Variance is the square of standard deviation.

Utility, risk aversion and risk premium

Suppose a job hunter in the field of management consulting is faced with
the following two options. The first is a job offer from a large
multinational company that promises to pay $50,000 a year. The second
is an offer from a small but growing local company that promises to pay
$20,000 a year plus a hefty $60,000 in commission assuming that the job-
hunter meets a million dollar sales quota per year. Her assessment,
however, shows that there is a 50 per cent chance that she meets this
quota and 50 per cent that she does not. The expected value of this lottery
is:
E(X) = 0.5 ($60,000) + 0.5 (0) = $30,000 (4)

32
 E5 Managerial Economics

Based on this calculation, which job would she choose? If she chooses the
job with the large multinational company, she is guaranteed $50,000, and
if she accepts the job with the small local company, her expected income
would be $50,000 ($20,000 salary plus $30,000 in expected commission).
On the surface, it appears that both jobs offer the same income. However,
it is very likely that she would choose the 100 per cent salary job (certain
outcome) with the multinational company over the local company’s
(risky) commission job. The reason is that the seemingly more exciting
job with the local company may in fact end up paying her just $20,000, if
our job-hunter is unable to meet her sales quota. This suggests that most
people, faced with two alternative projects of equal expected value of
profit but different coefficients of variation or risk, will generally prefer
the less-risky project (the one with the smaller coefficient of variation).
While it is true that some managers may very well choose the more risky
project (risk seekers) and some may be indifferent to either choice (risk
neutral), most managers are risk averters. The reason is to be found in the
principle of diminishing marginal utility of money. The meaning of
diminishing, constant, and increasing marginal utility of money will be
explained with the aid of a reward structure that helps explain
transformation of dollar payoffs into a more meaningful measurement.
Utility is such a measurement, and it can be expressed in conceptual units
called utils. Although difficult to establish a standard util by which one
can perform a cardinal measurement of utility, it is nonetheless a useful
concept.

Risk and diminishing marginal utility

At this point, it is necessary to explain the relationship between risk and
utility in a formal manner. To do so, profit and loss must be measured in
terms of marginal utility rather than absolute dollar values. Marginal
utility is defined as the change in total utility that takes place when one
more unit of money is gained or lost.

The three ways in which utility may theoretically relate to income are
depicted in Figures 2-3, 2-4, and 2-5. These depict behaviour of different
types of investors when investment yield or income is increased by equal
increments. Money income or wealth is measured along the horizontal
axis while the utility or satisfaction of money (measured in utils) is
plotted along the vertical axis. Each curve represents utility as a function
of income, U = U(I), where U stands for utils and I for income. The slope
of each curve represents marginal utility, which is where our interest lies.

The most common behaviour, depicted in Figure 2-3, is a risk avoider

(averter). The reason for risk aversion is diminishing marginal utility. It
shows that with no investment there is no return. A given increment
($20,000) to income when income is low, zero, increases utility by 50
(vertical axis), U($20,000) = 50, whereas the same increment to income
when income is, say $80,000, increases utility by a smaller amount,
U($100,000) –U($80,000) = (98–90) = 8.

33
Module 2

Figure 2-3

If, therefore, the total utility of the money curve is concave (or facing
down), doubling money income less than doubles utility. This is the basic
explanation of risk and can be used to illustrate the behaviour of our job
hunter.

Remember that our job hunter’s salary at the established multinational

company is $50,000, and as depicted in Figure 2-3, the level utility
associated with this income, U($50,000), equals 78, point A. The job
hunter’s income at the less-established local company, however, is one of
the two cases. She either makes $20,000, in case she fails to make any
commission income, in which case the corresponding utility would be 50,
point B, or she makes $80,000, if she succeeds in meeting her quotas,
where the corresponding utility would be 90, point C. Therefore, the job-
hunter’s expected utility at the local company is the expected value of the
utility levels she could receive if she worked for the local company:
0.50 x U($20,000) + 0.50 x U($20,000 + $60,000)
= 0.5 x 50 + 0.5 x 90
= 70. (5)

This is depicted by point D in Figure 2-3.

The above analysis shows that although the local company offers the
same expected salary as the established multinational company, the job
hunter’s expected utility at the local company, 70, is lower than the utility
she would receive from the job with the multinational company, 78.
Thus we see that the utility from the 100 per cent salary job (risk-free) is
greater than the expected utility from a commission-based job (risky) with
equal expected value of income. Therefore, if our job-hunter’s personality
fits that represented by Figure 2-3, she will prefer the risk-free to the risky
job. This is the preference of a decision maker who is risk averse.

In Figure 2-4, the utility function is a straight line, implying that doubling
income doubles utility so that the marginal utility of money is constant.

34
 E5 Managerial Economics

The straight-line utility function characterises a person who is indifferent

to risk, for whom marginal utility of a dollar lost is equal to that of a
dollar gained.
Figure 2-4

Finally, in Figure 2-5, if the total utility of money curve is convex or

faced down, doubling income more than doubles utility, so that the
marginal utility of money income increases. This represents the case of
compulsive gamblers, who place higher utility on dollars won than dollars
lost. The more they win, the more important winning becomes.
Figure 2-5

Most individuals are risk averters because their marginal utility of money
diminishes, that is, they face a total utility curve that is concave or faces
down. To see why this is so, consider the offer to engage in a bet to win
$10,000 if a head turns up in the tossing of a coin or to lose $10,000 if a
tail comes up. The expected value of the money won or lost is
Expected value of money income = E(I)
= 0.5($10,000) + 0.5(-$10,000) = 0

35
Module 2

Risk premium
Aversion to risk by managers and investors is manifested in many ways.
The following are a few examples of how risk may be averted. Grade AA
bonds sell for a higher price than grade B bonds. Investors diversify either
by creating individual portfolios or by investing in mutual funds. People
deposit their money in government treasury bills at low rates of interest
rather than in bonds that may earn substantially more interest. And people
buy all kinds of casualty and life insurance.

Why, then, if investors are averse to risk, do they put their money into
common stocks, commodities, precious metals, collectibles, and other
risky investments? The answer is that they do not do so unless they
receive a risk premium. The investor wants to be compensated not only
for the use of his or her money, but also for the risk that it may be lost. In
other words, the investor demands a higher rate of return when risk is
involved.

To illustrate this, we recall the example of our job hunter in Figure 2-6
below. In that example we showed that the job hunter preferred the risk-
free job (with the multinational company) to the risky job (with the local
company).
Figure 2-6

Accordingly, the risk premium is the minimum payment (compensation)

to the risk-averse decision maker (our job hunter) to make her indifferent
between the risky and risk-free events. In order to find the risk premium
for our job seeker, let us ask: at what level of sure (risk-free) income
(with the multinational company) would the resulting level of utility be
equal to the expected utility of the risky (commission-based) income? In
Figure 2-6, the expected utility of the risky job that is expected to pay
$50,000 is 70, point D. Therefore, the risk-free income whose
corresponding level of utility is also equal to 70 has to be about $40,000,
at point E. Note that E and D correspond to the same level of utility.
Therefore, our job hunter would be indifferent between a $40,000 (all

36
 E5 Managerial Economics

salary) job with the multinational firm and a $50,000 risky (partly
commission-based income) job with the local company. Hence, the risk
premium of the risky local company’s offer is $10,000.

Demonstration problem
Suppose you have a utility function that is concave (faced down). Also
suppose that you bet $100 on the flip of a coin at even odds. The
probability of winning is 0.5 and the probability of losing is also 0.5. If
you win, you will get $100 and if you lose you will lose $100. Should you
take the bet?
Answer:

Note that if you win you get $100 and if you lose you pay $100.
We also know that if you win you gain fewer utils of utility than
you sacrifice utils if you lose $100. This is because of the shape
of the utility function, diminishing marginal utility. Since the
probability of winning or losing is the same, the expected value in
utils is inevitably negative: 0.5(utils gained) + 0.5(utils
sacrificed) <0. This is so because utils gained are fewer than utils
sacrificed. Clearly, the investor should not take this bet.

Demonstration problem
Suppose you have a utility function that is concave (faced down). Also
suppose that you bet $100 on the flip of a coin at even odds. The
probability of winning is 0.5 and the probability of losing is also 0.5. This
time if you win, you will get $120 and if you lose you will lose $100.
Should you take the bet?
Answer:

It depends on the shape of your utility function. The $20 premium

may or may not be sufficient to make you indifferent between the
two possibilities. It may take more or less than $20 to take this
bet. The curvature of the utility function speaks to this issue. The
steeper the utility curve, the smaller the required risk premium,
and vice versa.

Risk adjustment in decision-making

In estimating the payoffs for a particular strategy, the decision maker
must consider both the present value of future returns and the degree of
risk. In this section we examine two of the most commonly used methods:
the risk-adjusted discount rate and the certainty-equivalent approach that
a risk-averse decision maker employs to compare decision alternatives on
a risk-adjusted basis.

37
Module 2

The risk-adjusted discount rate

Under conditions of risk and uncertainty the present value of future
returns are not known with certainty. Therefore, in estimating the payoffs
for a particular strategy, the decision maker needs to maximise the
(expected) Net Present Value (NPV), which combines the present value
calculation with expected-value analysis.
n
Rt
NPV    I0 (6)
t 1 (1  i )
t

where Rt represents the expected net return (cash flow) in each of the n
years considered, and i, as discussed in Module 1, is the appropriate
discount rate, and I0 is the amount of the initial investment.

One popular method of adjusting the NPV criterion of equation (6) to deal
with an investment project subject to risk is using higher discount rates
for more risky decision alternatives. We may define the risk-adjusted
discount rate as the required rate of return from a proposed investment
after due consideration of the risk involved:
n
Rt
NPV    I0 (7)
t 1 (1  r )
t

where r is the risk-adjusted discount rate, r = i + risk premium. As

discussed earlier, every firm has a required rate of return reflecting its
perception of its normal risk (normal business risk plus financial risk).

Demonstration problem
For example, suppose a firm’s normal business and financial risk requires
a 20 per cent rate of return. The firm is considering an investment strategy
that initially costs $100,000 and is expected to yield $50,000 cash inflow
per year for the next three years.
a. Calculate the net present value of the investment at a discount
rate of 20 per cent. Should the firm accept this project?
b. Suppose that the risk were such that management feels it should
get a 25 per cent return. Calculate the NVP for the adjusted
discount rate. Should the firm accept the investment project?
Answer:

a.

$50,000 $50,000 $50,000

NVP     $100,000  $5,324.
(1.20) (1.2) 2 (1.2) 3

38
 E5 Managerial Economics

Since the net present value is positive, a risk-neutral firm should

accept this project. A risk-averse firm may not necessarily accept
this project. It depends on the firm’s degree of risk-aversion.

b.

$50,000 $50,000 $50,000

NVP     $100,000  $  2,400.
(1.25) (1.25) 2 (1.25) 3

Here the NVP is negative. The project fails to provide a 25 per

cent discounted return and should be rejected by both a risk-
averse and a risk-neutral firm.

Thus we see that in the risk-adjusted discount-rate approach to evaluation

of proposed investments, risk is wholly reflected by the discount rate and
discounting process. There are, however, at least three limitations to this
approach to incorporation of risk:
1. How do we determine the appropriate discount rate? Clearly, the
introduction of a new product is riskier than buying government
bonds, but how much riskier? It is very difficult to resolve this
question consistently and objectively, particularly when there is no
historical evidence on which to base an estimate.
2. This method does not consider the probability distribution of future
cash flows information that could be of great value.
3. The risk-adjusted discount rate does not offer any consistent method
for evaluation of risk, an evaluation that may be quite subjective. This
limitation may be overcome by the certainty-equivalent approach.

The certainty-equivalent approach

The risk-adjusted discount-rate approach discussed in the preceding
section accounts for risk by simply modifying the discount rate appearing
in the denominator of the valuation model. In contrast, the certainty-
equivalent approach accounts for risk in the numerator of the valuation
model and uses a risk-free discount rate, i (such as the rate of return on
government bonds) in the denominator to account for the time value of
money. The degree of risk is reflected in the numerator by multiplying the
expected risky return by a certainty-equivalent coefficient.
n
 t Rt
NPV    I0 (8)
t 1 (1  r ) t

where  is the certainty equivalent coefficient, a number between 0 and 1.

1 means the project is risk free, and 0 means too risky to be considered.

Certainty equivalent of a decision alternative is the sum of money

available with certainty that would make the manager indifferent between
taking that decision and accepting the certain sum of money:

39
Module 2

R  R * (9)

where R* is the risk-free equivalent cash flow.

Demonstration problem
The manager of a company regards the sum of $80,000 with certainty as
equivalent to the expected (risky) net cash flow or return of $100,000 per
year for the next three years, what is the value of ?
Answer:

80,000
  0 .8
100,000

Asymmetric information
Asymmetric information refers to situations in which one party knows
more about its own actions or personal characteristics than another party.
When some people in the market have better information than others, the
people with the least information may choose not to participate in a
market. For example, the market for used cars is characterised by
information asymmetry. The seller always knows better than the buyer
about the quality of the car, including whether the car has been regularly
serviced and inspected or whether it has been involved in an accident.
That is why some people tend to shy away from used cars, unless some
form of warranty is tacked on.

Adverse selection
In the above example, while warranties may reduce the financial cost of
owning a lemon, they do not eliminate the bother, such as the time it takes
to bring the car into the shop. Of course, the owners know they have a
lemon and would like to pass it along to someone else. Those with the
worst lemons are going to be the most willing to sell their car, whatever
the price. But at a high used-car price, they will be joined by owners of
better-quality cars. As the price drops, more of the good cars will be
withdrawn from the market as the owners decide to keep them. And the
average quality of the used cars for sale will drop. We say there is an
adverse selection effect. The mix of those who elect to sell changes
adversely as price falls.

Asymmetric information affects many other managerial decisions,

including car insurance, employing workers, and issuing credit to
customers. Private car insurers do not have complete information about
the risk level of their clients, but they know that, as a group, young single
males demonstrate the highest claim frequency. The problem of adverse
selection arises because the insurers cannot identify the individuals who
are high risk. One way to rectify this matter, at least partially, is to

40
 E5 Managerial Economics

segment the market by age or gender charging higher premiums to high-

risk individuals.

This type of discriminating policy, however, is bound to be an imperfect

way to overcome the adverse selection problem. There will undoubtedly
be some drivers from certain categories, say, young males, who are
cautious drivers but who have to pay high premiums because of the
accident record of others of their gender and age group. On the other
hand, a non-discriminatory policy, charging the same premium to all
drivers will also discriminate between good and bad drivers by not
rewarding the good drivers. The end result is that some drivers are less
insured than they would like to be, because of high prices.

Another area where asymmetric information affects managerial decisions

is the job market. Job applicants have much better information about their
own capabilities than does the person in charge of hiring new workers. A
job applicant who claims to have excellent skills may be lying or not
telling the whole truth; the personnel manager has less information than
the applicant. This is why firms spend considerable amounts of time and
money setting up several interviews, designing tests to evaluate job
applicants, doing background checks, and the like. The basic reason for
these types of expenditures is to provide the firm with better information
about the capabilities and tendencies of job applicants.

Moral hazard
A second problem faced by insurance companies is an incentive problem.
Insurance reduces people’s incentives to attempt to avoid a loss and
encourages them to take excessive risk. If there existed some form of
business bankruptcy insurance, entrepreneurs would take riskier steps
than necessary or warranted. However, people buy insurance for their
house, cars and valuable belongings. For example, a person who has no
fire insurance on a house may choose to limit the risk by buying smoke
alarms and home fire extinguishers, and by being especially cautious.
However, if he has fire insurance, he might not be so as careful.
Therefore, moral hazard generally occurs when one takes hidden actions
that one knows another party cannot observe.

When moral hazard problems are strong, insurance firms will offer
limited or even no insurance. The limitations often take one of two forms:
deductibility provisions and co-insurance. Insurance policies may pay
damages only above some initial amount, referred to as a deductible. For
example, your car insurance policy may require you to pay the first $500
of damages before insurance benefits kick in. This reduces the moral
hazard problem associated with small claims; drivers might be much
more vigilant about avoiding minor accidents than major ones.
Alternatively, insurance policies may pay only some specified proportion
of damages. This is referred to as co-insurance. It forces those who are
insured to bear some cost of any accident and so to behave with greater
care.

41
Module 2

Module summary
This module has dealt with methods and approaches to decision-making
under conditions of risk and uncertainty. Under the condition of risk, the
primary decision criterion for selecting the optimum strategy is expected
value. The degree of risk is indicated by the standard deviation.
Summary
How decision makers choose to deal with risk depends upon their
attitudes. Some may try to seek risk, some may be indifferent toward it,
but most business people try to avoid risk. Their attitudes are based upon
utility functions in which increasing increments of income (profits) bring
decreasing increments of satisfaction (utility).

Risk aversion is based on the principle of diminishing marginal utility of

money, which is reflected in a total utility of money curve that is concave
or face down. A risk-averse decision maker will accept risk only if there
is a commensurate risk premium. Every business firm and individual
investor has in mind some required rate of return that reflects the
perceived risk. As the degree of risk increases, the required rate of return
also increases along a market-indifference curve that depicts the
investor’s risk-return trade-off function.

The profit-maximisation model can be simultaneously adjusted for both

risk and the true value of money by several techniques. Two of the most
common are the risk-adjusted discount rate and the certainty-equivalent
approach. The former involves adding a risk premium to the risk-free rate
of interest, or discount, used to find the present value of the net cash flow
or the return of the investment. A better method is the latter, which uses a
risk-free discount rate in the denominator and incorporates risk by
multiplying the net cash flow or return in the numerator of the valuation
by the certainty-equivalent coefficient.

42
 E5 Managerial Economics

Assignment
1. Suppose the required rate of return by a firm is 20 per cent while the
risk on government Treasury bills is 8 per cent. The firm is
considering an investment of $500,000 in a venture that promises to
yield $150,000 per year for the next five years.
a. Calculate the NPV of the proposed venture by the risk
Assignment adjusted discount method.
b. Calculate coefficient of a. that will cause the certainty-
equivalent approach to yield the same result.
2. Suppose a firm is considering an investment of $100,000 that is
expected to yield a cash flow of $50,000 per year for three years.
Suppose that management’s perception of risk is such that it
considers risk-free returns of $45,000 in the first year, $40,000 in the
second year, and $35,000 in the third year to be equivalent to the
risky return of $50,000 for each year. Calculate NPV for each of the
three years.

43
Module 2

Assessment
1. Why is the certainty-equivalent approach to risk adjustment
considered to be superior to the risk-adjusted approach?
2. Explain the concept of risk premium. What causes the magnitude of
Assessment
risk premium change?
3. ‘For a risk-averse consumer the expected utility of a gamble is greater
than the utility of the expected value of that gamble.’ True or False?
Explain.
4. ‘For a risk-lover consumer the expected utility of a gamble is greater
than the utility of the expected value of that gamble.’ True or False?
Explain.
5. Explain why a used car that is only six months old and has been
driven only 5,000 km typically sells for 20 per cent less than a new
car with the same options.

44
 E5 Managerial Economics

Assessment answers
1. If there are any time periods in which perceived risk is more or less
than what is represented by the risk-adjusted discount rate, then the
certainty-equivalent approach can provide a better estimate of the net
present value of a proposed investment. For example, the return from
an investment for the introduction of a new product might be more
uncertain in its earlier years, while the firm is struggling for product
recognition and market share, than in later years when the product’s
market has become established. The certainty equivalent approach
can easily handle this situation during the process of establishing 
for each separate time period. The certainty-equivalent approach
enables managers to specify directly the degree of risk for and then
discount the cash flow.

2. The risk premium is the minimum payment (compensation) to the

risk averse decision maker (our job hunter) to make her indifferent
between the risky and risk-free events. The magnitude of risk
premium changes with changes in attitude towards risk. The higher
the degree of risk aversion the greater the risk premium required by
the decision maker.

3. FALSE: If two events (one risky and one safe) have the same
outcome, due to diminishing marginal utility of income, the expected
utility of the risky event will always be smaller than the utility of the
safe asset (concave utility function). You can show this by drawing
the marginal utility of income curve under alternative risk attitudes.

4. TRUE. See answer to question 3.

5. This is due to asymmetric information. The buyer of the used car is

not aware of the behaviour and driving habits of the first owner –
adverse selection (hidden characteristics). The first owner knows how
the car has been kept up, while the buyer does not.

45
Module 2

References
Arrow, K. J. (1982). Risk Perception in Psychology. Economic Inquiry.
Baye, M. (2002). Managerial Economics and Business Strategy. Irwin:
McGraw Hill.
References
Graham, D. (1981). Cost- Benefit Analysis under Uncertainty. American
Economic Review, September.

46