TCHE341:

FINANCIAL ECONOMICS

CHAPTER 3: Theory of choice: utility

theory given uncertainty
Đỗ Khánh Hiền
 Theory of choice: utility theory given uncertainty

CHAPTER 3: OUTLINE

❖ Five axioms of choice under uncertainty

❖ Developing utility functions
❖ Establishing a definition of risk aversion and choice
criteria
❖ Comparison of risk aversion in the small and in the large
❖ Stochastic dominance
❖ Using mean and variance as choice criteria
 Theory of choice: utility theory given uncertainty

LECTURE 5: OUTLINE

❖ Five axioms of choice under uncertainty

❖ Developing utility functions
❖ Objectives: Five axioms of choice under
uncertainty; Utility functions
 Theory of choice: utility theory given uncertainty

Introduction

Theory of
choice
Theory of
optimal decision
making under
uncertainty

Objects of
choice
Price
s?
 Theory of choice: utility theory given uncertainty

Utility theory
 Theory of choice: utility theory given uncertainty

Five axioms of choice under uncertainty

➢ Axiom 1: Comparability (sometimes called completeness)

For the entire set, S, of uncertain alternatives, an individual can say
either that outcome x is preferred to outcome y (we write this x y) or
y is preferred to x (y x) or the individual is indifferent as to x and y
(x ~ y)
 Theory of choice: utility theory given uncertainty

Five axioms of choice under uncertainty

➢ Axiom 2: Transitivity (sometimes called consistency)

If an individual prefers x to y and y to z, then x is preferred to z (If x
y, y z then x z). If an individual is indifferent as to x and y and is
also indifferent as to y and z, then he or she is indifferent as to x and z
(x ~ y, y ~ z then x ~ z)
 Theory of choice: utility theory given uncertainty

Five axioms of choice under uncertainty

➢ Axiom 3: Strong independence

If x ~ y then G(x, z: α) ~ G(y, z: α)
Suppose we construct a gamble where an individual has a probability α
of receiving outcome x and a probability (1 - α) of receiving outcome
z. We shall write this gamble as G(x, z: α). Strong independence says
that if the individual is indifferent as to x and y, then he or she will also
be indifferent as to a first gamble, set up between x with probability α
and a mutually exclusive outcome z, and a second gamble, set up
between y with probability α and the same mutually exclusive
outcome, z.
 Theory of choice: utility theory given uncertainty

Five axioms of choice under uncertainty

➢ Axiom 4: Measurability
If or
then there exists a unique α such that y ~ G(x, z: α)
If outcome y is preferred less than x but more than z, then there is a
unique α such that the individual will be indifferent between y and a
gamble between x with probability α and z with probability (1-α).
 Theory of choice: utility theory given uncertainty

Five axioms of choice under uncertainty

➢ Axiom 5: Ranking
If or then if y ~ G(x, z: α1)
and u ~ G(x, z: α2), it follows that if α1 > α2 then
or if α1 = α2 then y ~ u
If alternatives y and u both lie somewhere between x and z and we can
establish gambles such that an individual is indifferent between y and a
gamble between x (with probability α1) and z, while also indifferent
between u and a second gamble, this time between x (with probability
α2) and z, then if α1 is greater than α2, y is preferred to u.
 Theory of choice: utility theory given uncertainty

Assumptions about behavior

The axioms of cardinal utility based on certain assumptions
➢ First, all individuals are assumed to always make completely rational decisions.
A statement that "I like oranges more than apples and apples more than grapes but grapes more
than oranges" is not rational.
➢ Second, people are assumed to be able to make these rational choices among
thousands of alternatives
➢ The axiom of strong independence is usually the hardest to accept
Let outcome x be winning a left shoe, let y be a right shoe, and let z also be a right shoe.
Imagine two gambles: 1) 50/50 chance of winning x or z (left/right); 2) 50/50 chance of winning
y or z (right/right). If we were originally indifferent between the choice of a left shoe (by itself)
or a right shoe (by itself), then strong independence implies that we will also be indifferent
between the two gambles we constructed. Of course, left shoes and right shoes are
complementary goods, and we would naturally prefer to have both if possible. The point of
strong independence is that outcome z in the above examples is always mutually exclusive.
 Theory of choice: utility theory given uncertainty

Developing utility functions

The utility function will have two properties:

➢ Order preserving: If U(x) > U(y) then x is preferred to y
(or x y)
➢ Expected utility can be used to rank combinations of risky
alternatives
U[G(x, y: α)] = αU(x) + (1 - α)U(y)
 Theory of choice: utility theory given uncertainty

Developing utility functions

Proving Order preserving:

Consider the set of risky outcomes, S, assumed to be bounded above
by outcome a and below by outcome b. Next consider two intermediate
outcomes x and y such that

By using Axiom 4 (measurability), we can choose unique probabilities

for x and y in order to construct the following gambles:
 Theory of choice: utility theory given uncertainty

Developing utility functions

According to Axiom 5 (ranking): probabilities α(x) and α(y) can be

interpreted as numerical utilities that uniquely rank x and y:
If α(x) > α(y) then
If α(x) = α(y) then x ~ y
If α(x) < α(y) then
 Theory of choice: utility theory given uncertainty

Developing utility functions

Show that expected utility can be used to rank risky alternatives.

 Theory of choice: utility theory given uncertainty

Developing utility functions

Next, consider a third alternative, z. According to Axiom 3 (strong

independence) z will not affect the relationship between x and y.
By Axiom 4, there must exist a unique probability β(z) that would make us
indifferent as to outcome z and a gamble between x and y
 Theory of choice: utility theory given uncertainty

Developing utility functions

Relating z to the elemental prospects:

With U(x) = α(x) & U(y) = α(y):

Using axioms 4 & 5:
Generally:
 Theory of choice: utility theory given uncertainty

Developing utility functions

Note: Utility functions are individual-specific => There is no way to compare
one individual's utility function to another's!
Suppose we arbitrarily assign a utility of -10U to a loss of $1000
(or U(-1000) = -10).
When we are faced with a gamble with probability α of winning $1000 and
probability (1- α) of losing $1000;
What probability would make us indifferent between the gamble and $0.0
with certainty, or and
Assuming α = 0.6 then
 Theory of choice: utility theory given uncertainty

Developing utility functions

 Theory of choice: utility theory given uncertainty

Developing utility functions

 Theory of choice: utility theory given uncertainty

Increasing or decreasing marginal utility

Consider the freezing point (y) & boiling point (x) of water in C and F scale:
Each scale may be likened to a function that maps various degrees of heat into
numbers. Utility functions do the same thing for risky alternatives.
The difference between two outcomes is marginal utility.
The ratio of the "changes" is
If the two scales really do provide the same ranking for all prospects, then the
ratio of changes should be the same for all prospects:

Changes in utility between any two wealth levels have exactly the same meaning
on the two utility functions, i.e., one utility function is just a "transformation" of
the other
 Theory of choice: utility theory given uncertainty

LECTURE 6: OUTLINE

❖ Establishing a definition of risk aversion and choice

criteria
❖ Comparison of risk aversion in the small and in the
large
❖ Stochastic dominance
❖ Using mean and variance as choice criteria
❖ Objectives: Risk aversion; Choice criteria;
Stochastic dominance
 Theory of choice: utility theory given uncertainty

Risk Aversion
More wealth is preferred to less!
➢ In other words: MU(W) > 0 (the marginal utility of wealth is
positive)
➢ Suppose that we establish a gamble between two prospects G(a, b: α)
➢ Will we prefer the actuarial value of the gamble (i.e., its expected or
average outcome) with certainty or the gamble itself?
➢ Would we like to receive $10 for sure or prefer to "roll the dice" in a
gamble that pays off $100 with a 10% probability and $0 with a 90%
probability?
➢ Depending on one’s risk tolerance/appetite!
 Theory of choice: utility theory given uncertainty

Risk Aversion
 Theory of choice: utility theory given uncertainty

Risk Aversion

Utility function: U(W) = ln(W)

Example: G(5; 30; 80%)
➢ E(W) = $5*80% + $30*20% = $10
➢ U[E(W)] = 2.3
➢ If an individual with a logarithmic utility function could receive
$10 with certainty, it would provide him or her with 2.3 utiles
 Theory of choice: utility theory given uncertainty

Utility function
 Theory of choice: utility theory given uncertainty

Risk Aversion
➢ If U[E(W)] > E[U(W)] => risk aversion
➢ If U[E(W)] = E[U(W)] => risk neutrality
➢ If U[E(W)] < E[U(W)] => risk loving
➢ Note: If our utility function is strictly concave, linear or convex we will
be risk averse, risk neutral and risk lovers (respectively)
 Theory of choice: utility theory given uncertainty

Risk Aversion

➢ Risk premium: the maximum amount of wealth an individual

would be willing to give up in order to avoid the gamble.
➢ Gamble G(5;30;80%) has:
E[U(W)] = 1.97, meaning W = 7.17
➢ To avoid the gamble (with E(W) = $10) , individual A is willing
to pay: Markowitz risk premium = $10 - $7.17 =$2.83
➢ Certainty equivalent wealth: the level of wealth that individual
would accept with certainty if the gamble were removed
➢ Risk premium = E(W) - CEW
CE

lottery ticket: p =50%

win $20k or get nothing ($0) = > EV = $10k
CE: lowest amount of money-for-certain that the decision maker would be willing to
accept instead of a gamble (eg: 7k indifferent lottery vs cash 7k)
RP = 3k

G’(10;100;10%): E(W) = $101, CEW = $92.76

E(W) = 10*10%+ 100*90% + 10 = $101
cew: E(U(W)) = 10%U(10+10)+90%U(100+10)
= 10%*U(20)+90%*U(110) = 4.53 = > CEW= 92.76
ln(x) = 4.53 = > x
 Theory of choice: utility theory given uncertainty

Utility function
 Theory of choice: utility theory given uncertainty

Risk Aversion

➢ Risk premium vs cost of the gamble

➢ Assuming individual B has logarithmic utility function lnU(W)
and CW = $10
➢ G’(10;100;10%) has E(W) = $101, CEW = $92.76
➢ To avoid G’, B is willing to pay:
Markowitz risk premium = $101 - $92.76 =$8.24
➢ Note: E(W) > CW => the gamble is favorable
➢ Cost of the gamble = CW - CEW = -$82.76 => B willing to pay
$82.76 to take G’
 Theory of choice: utility theory given uncertainty

Risk Aversion
➢ We shall assume that all individuals are risk averse with strictly
concave and increasing utility functions.
➢ Individual T with a current amount of wealth W is offered an
actuarially neutral gamble of Z dollars (E(Z) = 0)
➢ What risk premium π must be added to the gamble to make her
indifferent between it and the actuarial value of the gamble?

➢ The left-hand side is the expected utility of the current level of

wealth, given the gamble
➢ The right-hand side: current level of wealth + utility of the
actuarial value of the gamble - risk premium
 Theory of choice: utility theory given uncertainty

Risk Aversion

➢ Taylor’s series approximation of the left-hand side:

➢ Note:

➢ Equate the equations:

 Theory of choice: utility theory given uncertainty

Risk Aversion
➢ Pratt-Arrow measure of a local risk premium

➢ The sign of the risk premium is determined by that of the term in parentheses
➢ Absolute risk aversion:
ARA measures risk aversion for a given level of wealth.

➢ Relative risk aversion:

RRA = W x ARA
TEST: ARA decreases, Constant relative risk aversion implies that an individual
will have constant risk aversion to a proportional loss of wealth even though the
absolute loss increases as wealth does?!
 Theory of choice: utility theory given uncertainty

Risk Aversion

➢ Friend & Blume (1975) used sophisticated econometric techniques to

estimate changes in ARA and RRA as a function of the wealth of
investors.

➢ Results:

➢ The marginal utility of wealth is positive and decreases with

increasing wealth.
➢ ARA decreases with increasing wealth, and RRA is constant.
 Theory of choice: utility theory given uncertainty

Comparison Of Risk Aversion In The Small

And In The Large

➢ Pratt-Arrow definition of risk aversion assumes that risks are

small and actuarially neutral
➢ The Markowitz concept, which simply compares E[U(W)] with
U[E(W)] is not limited by these assumptions.
➢ Ex: An individual with a logarithmic utility function and a level
of wealth of $20k is exposed to two different risks:
(1) a 50/50 chance of gaining or losing $10
(2) an 80% chance of losing $1,000 and a 20% chance of losing
$10k
 Theory of choice: utility theory given uncertainty

Risk (1)
According to Pratt-Arrow:

Calculations:
 Theory of choice: utility theory given uncertainty

Risk (1)

According to Markowitz:

Risk premium RP = $.0025002

 Theory of choice: utility theory given uncertainty

Risk (2)

The Pratt-Arrow assumptions of a small, actuarially neutral risk are not

closely approximated.
Nevertheless, if we apply the Pratt-Arrow definition, the risk premium is
calculated to be RP = $324
Markowitz: RP = EW - CEW = $17,200 - $16,711 = $489
 Theory of choice: utility theory given uncertainty

Stochastic Dominance

➢ An asset (or portfolio) is said to be stochastically dominant over

another if an individual receives greater wealth from it in every
(ordered) state of nature (first-order stochastic dominance)
➢ Asset x, with cumulative probability distribution Fx(W) will be
stochastically dominant over asset y, with cumulative
probability distribution Gy(W) for the set of all nondecreasing
utility functions if
 Theory of choice: utility theory given uncertainty

Stochastic Dominance

➢ In other words, the cumulative probability distribution (defined

on wealth, W) for asset y always lies to the left of the
cumulative distribution for x. If true, then x is said to dominate
y.
 Theory of choice: utility theory given uncertainty

Using Mean And Variance As Choice Criteria

➢ If the distribution of returns offered by assets is jointly normal,

then we can maximize expected utility simply by selecting the
best combination of mean and variance.
➢ The relationship between wealth and return:

➢ If the end-of-period wealth from investing in asset j is normally

distributed with mean and variance
(1) mean E(Rj) =
(2) Variance
 Theory of choice: utility theory given uncertainty

Indifference curves for a risk-averse investor

➢ Assuming that the return on an asset is normally distributed with

mean E and variance , can write our utility function as

➢ Expected utility:

➢ The indifference curve is a mapping of all combinations of risk and

return (standard deviation or variance) that yield the same expected
utility of wealth.
 Theory of choice: utility theory given uncertainty

Recent Thinking And Empirical Evidence

➢ Utility theory is founded on the axioms of Von Neumann and Morgenstern
[1947] and the elegant mathematics which follows logically from them.
➢ There has been almost no empirical testing of the axioms or of their
implications, at least not by economists.
➢ Psychologists, however, have been busy testing the validity of the axioms.
The answer seems to be that individuals actually do not behave as
described by the axioms.
➢ Kahneman and Tversky [1979, 1986] point out that the way decisions are
framed seems to matter for individual decision making. They give an
example where people are asked to decide between surgery and radiation
therapy for cancer treatment. The information in both frames is exactly the
same, yet when presented with the survival frame, 18 percent preferred
radiation, and when presented with the mortality frame, 44 percent
preferred radiation.
 Theory of choice: utility theory given uncertainty

Kahneman & Tversky [1979, 1986]

➢ Survival Frame
Surgery: Of 100 people having surgery, 90 live through the postoperative
period, 68 are alive at the end of the first year, and 34 are alive at the end of
five years.
Radiation Therapy: Of 100 people having radiation therapy, all live through
the treatment, 77 are alive at the end of one year, and 22 are alive at the end
of five years.
➢ Mortality Frame
Surgery: Of 100 people having surgery, 10 die during surgery or the
postoperative period, 32 die by the end of the first year, and 66 die by the
end of five years.
Radiation Therapy: Of 100 people having radiation therapy, none die
during treatment, 23 die by the end of one year, and 78 die by the end of
five years.
 Theory of choice: utility theory given uncertainty

Exercises

1, A small businessman faces a 10% chance of having a fire that will

reduce his net worth to $1.00, a 10% chance that fire will reduce it to
$50,000, and an 80% chance that nothing detrimental will happen, so
that his business will retain its worth of $100,000.
What is the maximum amount he will pay for insurance if he has a
logarithmic utility function? In other words, if U(W) = ln W, compute
the cost of the gamble. [Note: The insurance pays $99,999 in the first
case; $50,000 in the second; and nothing in the third.]
 Theory of choice: utility theory given uncertainty

Exercises
2, If you are exposed to a 50/50 chance of gaining or losing $1000 and
insurance that removes the risk costs $500, at what level of wealth will
you be indifferent relative to taking the gamble or paying the
insurance? That is, what is your certainty equivalent wealth? Assume
your utility function is
3, Roberts Company has a cash inflow of $140,000 per year for project
A and an outflow of $100,000 per year. The investment cost in the
project is $100,000; its service life is 10 years; The tax rate is 40%.
The opportunity cost of capital is 12%.
a, Calculate the NPV of project A, using straight-line depreciation for
tax purposes
b, Assume EBITDA is $22,000 per year. Calculate the NPV of A?