5QQMB204 INTERMEDIATE MICROECONOMICS

LECTURE 7: CHOICE UNDER RISK AND UNCERTAINTY

• Perloff, Chapters 17
Recall: Consumer Choice and Demand Under Certainty
• Choosing the optimal bundle/combination of two goods
• Each combination is expected to give the same level of
satisfaction/utility.
• Indifference curve analysis
• Shows combinations of two goods that give the consumer the same
level of satisfaction/utility.
• ICs further out give higher levels of satisfaction.
• Shape of IC
• Downward sloping (for normal goods) implies a diminishing marginal
rate of substitution (MRS) due to diminishing marginal utility.

2
Indifference curve analysis

3
Finding the optimum consumption

4
Recall: Consumer choice and demand under certainty

• But these analysis have some limitations:

• Consumers may face uncertain choices.
• Consumers may not behave rationally.

5
 Uncertainty and risk
• Uncertainty affects consumption decisions.
• Consumers consider the outcomes of their choices under a
number of possible circumstances (states of nature) – rejection
or acceptance.
• Risk is sometimes used to quantify these uncertainties.
• What is meant by risk?
• Hazard or chance of loss
• i.e.; a situation in which the likelihood of each possible
outcome is known or can be estimated, and no single
possible outcome is certain to occur
• How do people decide what to do in uncertainty?

6
 Uncertainty and risk

• Consumers have some idea of which outcomes are more

likely than others.
• Consumers modify their decisions as the degree of risk
varies.
• Choose a riskier option if they receive a higher payout.
• Probabilities are used to quantify how risky outcomes are
• Probabilities are used to determine expected earnings.

7
Assessing risk - Probability
• Probability indicates how likely an event is to occur
• Ranges between 0 and 1
• The sum of the probabilities of all possible outcomes is equal to 1
• We use the frequency with which an outcome occurred to estimate
the probability
• For instance, if it rained 20 times on an August bank holiday in the last 40
years, then the frequency that is rained, 𝜃 theta, is given by;
!
• 𝜃="
#$
• 𝜃= %$
= 0.5
Probability

• But often, there is no history of repeated events –

because they happen infrequently
• So, we are unable to calculate frequencies
• Then, we use subjective probability
• The subjective probability can combine frequencies and all
other available information.
• Even information that is not based on scientific
observation.
• And sometimes, the estimates we make are wrong, e.g. if
we give too much weight to unlikely events.

9
Probability Distribution
• A probability distribution relates the probability of
occurrence to each possible outcome.
• Mutually exclusive—when only one of the outcomes can
occur at a given time.
• Exhaustive—when no other outcomes than those listed
are possible.
• Where outcomes are mutually exclusive and
exhaustive, exactly one of these outcomes will occur
with certainty, and the probabilities must add up to
100%.
Probability Distribution
 Expected value

• This is the amount an individual expects to receive (payoff)

for a transaction
• Assume that there are two outcomes, X and Y;
• EV = Pr(X)*V(X) + Pr(Y)*V(Y)
• Where Pr() is the probability and V() is the value of the
outcome.

12
 Expected value - example
• Gregg, a promoter, schedules an outdoor concert for tomorrow.
• How much money he’ll make depends on the weather.
• If it doesn’t rain, his profit or value from the concert is V =
$15.
• If it rains, he’ll have to cancel the concert and he’ll lose V =
−$5, which he must pay the band.
• He knows that the weather department forecasts a 50% chance
of rain.
• Expected Value?
𝐸𝑉 = Pr(𝑛𝑜 𝑟𝑎𝑖𝑛)×𝑉𝑎𝑙𝑢𝑒(𝑛𝑜 𝑟𝑎𝑖𝑛) + Pr(𝑟𝑎𝑖𝑛)×𝑉𝑎𝑙𝑢𝑒(𝑟𝑎𝑖𝑛)
1 1
= ×$15 + ×(−$5) = $5
2 2

13
 Variance and standard deviation

• EV tells us the likely payout – the expected value of the payout.

• Gregg cannot tell the risk simply by using the EV
• how can we determine risk?
• Variance (standard deviation) is used:
• Variance (standard deviation): Shows the spread of the probability
distribution; how close together are the possible outcomes.
• Var =[Pr(X) * (V(X) – EV)2] + [ Pr(Y) * (V(Y) – EV)2]
• SD (σ) is square root of Var
• The greater the SD relative to the EV, the greater the risk.

14
 Variance and standard deviation

• Example: Probability it rain = 0.5; value of outcome if it does not

rain = 15; value of outcome if it rain = -5 Therefore EV = 5
• 𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒 = Pr(𝑛𝑜 𝑟𝑎𝑖𝑛)×(𝑉𝑎𝑙𝑢𝑒 𝑛𝑜 𝑟𝑎𝑖𝑛 − 𝐸𝑉)# + [Pr 𝑟𝑎𝑖𝑛 ×
𝑉𝑎𝑙𝑢𝑒 𝑟𝑎𝑖𝑛 − 𝐸𝑉 #]
& &
• 𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒 = ×(15 − 5)# + ×((−5) − 5)#
# #
& # & #
• 𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒 = #
×(10) + #
×(−10) = 100
• Standard deviation (σ)- the square root of the variance – Hence
Standard deviation = 10

15
Example: Deal or no deal

16
 Deal or no deal – example 1

• How it works: 22 boxes with varying amounts. The contestant is

allocated one box randomly. In each round, three boxes are opened,
and the contestant gets an offer of an amount from the banker to
quit the game.
• Suppose the contestant has the following amount left in the boxes.
£1000
£5
£10000
£500 £50000

• The banker offers the contestant £6500 to abandon the game. What
should the contestant do?
• What is the expected value?

17
 Deal or no deal
• What is the expected value?
• Each outcome is equally likely. Hence, the probability of any
outcome =1/5
• 𝐸𝑉 = 0.2 £5 + 0.2 £500 + 0.2 £1000 + 0.2 £10000 +
0.2 £50000 = £12301
• The banker’s offer (£6500) is below the expected value
• The difference (£12301- £6500 = £5801) is the risk premium
• The amount the consumer will pay to avoid the risk
• If the individual chooses the banker’s offer they are willing to
pay the risk premium to avoid the risk of the gamble.

18
 Deal or no deal - example 2

• Now, suppose the contestant has the following amount

left in boxes
£5 £1000

£50 £250000

£100

• The banker offers the contestant £12500 to abandon

the game.
• The expected value =?

19
 Deal or no deal - example 2

• Now, suppose the contestant has the following amount

left in boxes
£5 £1000

£50 £250000

£100

• The banker offers the contestant £12500 to abandon

the game.
• The expected value = 0.2(5) + 0.2(50) + 0.2(100) +
0.2(1000) + 0.2(250000) = 50231

20
 Deal or no deal – which gamble is riskier?
• The greater the SD relative to the EV, the greater the risk.
• The banker’s offer relative to the EV:
• In the first example, this is £6500/£12301 = 53%
• In the second example: £12500/£50231 = 25% of the EV
• Standard deviation in the first example = £19208
• Risk is, therefore, £19208/£12301 = 1.56
• Standard deviation in the second example = £99825
• Risk = £99825/£ 50231 = 1.99
• The second game is riskier than the first

21
 Attitudes towards risk

• if individuals do not care about risks, they will choose the

option with the highest expected value.
• But most people are risk averse – and so are willing to pay a
premium to avoid risk, especially if an option is particularly
risky.
• How can you determine if the EV of a risker option is
sufficiently higher to justify the greater risk?
• Extension of model of utility maximisation

22
 Expected utility
• We assume that the individual knows the value of each possible
outcome and the probability that it will occur.
• A rational person maximises expected utility.
• Expected utility (EU) is the probability-weighted average of the utility
from each possible outcome.
• 𝐸𝑈 = Pr 𝑋 ×𝑈 𝑋 + Pr 𝑌 ×𝑈 𝑌
• Where the utility function U depends on payoffs; U(X) is the utility
from X; U(Y) is the utility from Y
• EU captures the trade-off between risk and value, whereas the
expected value considers only value.

23
 Expected utility

• For example, Gregg’s EU from the outdoor concert is:

• EU= Pr(𝑛𝑜 𝑟𝑎𝑖𝑛)×𝑈(𝑉𝑎𝑙𝑢𝑒 𝑛𝑜 𝑟𝑎𝑖𝑛 ) + [Pr(𝑟𝑎𝑖𝑛)×
𝑈(𝑉𝑎𝑙𝑢𝑒 𝑟𝑎𝑖𝑛 )]
! !
• 𝐸𝑈 = "
×𝑈($15) + "
×𝑈(−$5)
• EU depends on the individual’s utility for money in this
case
• If we know how an individual’s utility increases with
wealth, we can determine how that person reacts to risky
situations.

24
 Expected utility

• Risk-averse - prefer a choice with a more certain outcome to

one with a less certain outcome
• Utility function has diminishing marginal utility
• The extra utility they get from one more dollar is less that they got
from the last
• Risk lovers – prefer a gamble with a less certain outcome to
one with a more certain outcome.
• because they are willing to give up some expected return to take on
more risk, the utility function has increasing marginal utility.
• Risk neutral – They maximise expected wealth, regardless of
risk.
• Will choose whichever option has the higher rate of return because
they do not care about risk. The utility function is linear, and the
marginal utility is constant.

25
 Example: Risk aversion
• Irma, who is risk averse, makes a choice under uncertainty. She has
an initial wealth of $40 and has two options:
• Nothing, and keep the $40 (her utility is U(40)=120) with
certainty.
• Buy a share
• Her wealth will be:
• $70 if the start-up is a success – utility of 70, U(70) = 140
• $10 otherwise – utility of 10, U(10) = 70
• Irma’s subjective probability is 50% that the firm will be a success.
• Her expected utility from buying a share is:
• 0.5U(10) + 0.5U(70) = 0.5*70 + 0.5*140 = 105

26
Risk aversion

27
 Risk premium

• Risk premium - the amount that a risk-averse person would

pay to avoid taking a risk.

• The difference between the expected value of the uncertain

prospect and the certainty equivalent;
• 40 - 26 = 14

28
Risk neutrality and risk preference
constant increasing
marginal utility marginal utility of
of wealth, wealth

29
 Risk avoidance through risk sharing and
spreading

• There are a lot of ways people can avoid or decrease risk

• Insurance: a way of removing risks for the risk-averse
• Insurers make profits by spreading or sharing their risks
• the law of large numbers – risk pooling – profits on average
- as long as premium are high enough.
• Problems – risks - for insurers - adverse selection & moral
hazard
• Diversification can also reduce risk – portfolio-based investment.
• breaking risk into pieces (as long as gambles are not
correlated) = securitisation

30
 Investing Under Uncertainty
Attitudes toward risk affect people’s willingness to invest.

• Risk-Neutral Investing - Example

• Chris, the owner of the monopoly, is risk neutral.
• She maximizes her expected utility by making the investment only if the expected value of the
return from the investment is positive.
(a) Risk-Neutral Owner
 Investing Under Uncertainty
• Risk-Averse Investing – Example
• Ken, who is risk averse, faces the same decision as Chris.
• Ken invests in the new store if his expected utility from
investing is greater than his certain utility from not
investing.
(b) Risk-Averse Owner
 Behavioral Economics of Uncertainty

• Do people behave like we have just described?

• Economists and psychologists explain why most individuals make
choices under uncertainty that are inconsistent with the predictions
of expected utility theory.
• Biased Assessment of Probabilities
• Gambler’s fallacy—arises from the false belief that past events
affect current, independent outcomes.
• Overconfidence - gamblers are overconfident.
• The Certainty Effect Many people put excessive weight on
outcomes they consider to be certain relative to risky outcomes.
 Prospect Theory Value Function
• Prospect theory
• an alternative theory of decision-making under uncertainty that can
explain some of the choices people make that are inconsistent with
the expected utility theory.