Overview of Chapter 15
In this lecture, we will cover:
The axiomatic foundations of utility theory
Certain paradoxes from cognitive psychology that
affect decision making
Why and how people do not make choices that are
perfectly consistent with the axioms
• Yet they agree that the axioms are reasonable!
Implications and paradoxes of decision making
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Example
The United States is preparing for an outbreak of an unusual Asian strain of influenza. Experts
expect 600 people to die from the disease. Two programs are available that could be used to
combat the disease, but because of limited resources only one can be implemented.
Program A (Tried and True) 400 people will be saved.
Program B (Experimental) There is an 80% chance that 600 people will be saved and a 20%
chance that no one will be saved
.
Which of these two programs do you prefer?
Now consider the following two programs:
Program C 200 people will die.
Program D There is a 20% chance that 600 people will die and an 80% chance that no one
will die.
Would you prefer C or D?
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Axioms & Paradoxes
What is an Axiom?
An axiom is a statement or proposition which is
regarded as being established, accepted, or self-
evidently true. Similar words/terms:
• Accepted truth
• General truth
• Dictum
• Truism
• Principle
• Proposition
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Axioms & Paradoxes
What is a Paradox?
A paradox a statement or proposition that, despite
sound (or apparently sound) reasoning from
acceptable premises, leads to a conclusion that
seems senseless, logically unacceptable, or self-
contradictory. Similar words/terms:
• Contradiction • Inconsistency
• Contradiction in terms • Proposition
• Self-contradiction
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Axioms for Expected Utility
Axioms: Rules for clear thinking
From the following behavioral axioms, it is possible to
establish logically that people who behave according
to the axioms should make choices consistent with the
maximization of expected utility.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Axioms for Expected Utility
Axioms for Expected Utility
1. Ordering and Transitivity
2. Reduction of Compound Uncertain Events
3. Continuity
4. Substitutability
5. Monotonicity
6. Invariance
7. Finiteness
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Axioms for Expected Utility
1. Ordering and transitivity
Ordering: A decision maker can rank order
(establish preference or indifference) any two
consequences, and the ordering is transitive.
Transitivity: If A1 is preferred to A2 and A2 is
preferred to A3, then A1 is preferred to A3.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Axioms for Expected Utility
2. Reduction of compound uncertain events
Reduction axiom: A decision maker is indifferent
between a compound uncertain event (a
complicated mixture of gambles or lotteries) and a
simple uncertain event as determined by reduction
using standard probability manipulations.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Axioms for Expected Utility
3. Continuity
Given three consequences A1, A, and A2 where the
decision maker’s stated preferences are A1 > A > A2,
then we can always find a probability value p, 0 < p <1,
such that the decision maker is indifferent between
receiving A for sure and a lottery with probability p of
receiving A1 and probability 1 – p of receiving A2.
We can always construct a reference gamble with
some probability p, 0 < p <1, for which the decision
maker will be indifferent between the reference gamble
and A.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Axioms for Expected Utility
4. Substitutability
A decision maker is indifferent between any original
uncertain event that includes outcome A and one
formed by substituting for A an uncertain event that
is judged to be its equivalent.
Allows for the substitution of uncertain reference
gambles into a decision for their CE
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Axioms for Expected Utility
Two decision trees that show how substitutability works: If A is equivalent to a
lottery with a p chance at C and 1 – p chance at D, then Decision Tree I is equivalent to
Decision Tree II.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Axioms for Expected Utility
5. Monotonicity
Given two reference gambles with the same
possible outcomes, a decision maker prefers the
one with the higher probability of winning the
preferred outcome.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Axioms for Expected Utility
6. Invariance
All that is needed to determine a decision maker’s
preferences among uncertain events are the
payoffs (or consequences) and the associated
probabilities.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Axioms for Expected Utility
7. Finiteness
No consequences are considered infinitely bad or
infinitely good.
The finiteness axiom assures us that expected
utility will never be infinite, and so we always will be
able to make meaningful comparisons
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Axioms for Expected Utility
These axioms are needed to construct predictive or
descriptive Expected Utility models in Decision
Analysis.
They define a rational decision maker.
However, people do not always make rational
decisions that are in accordance with utility theory.
The situations can result in paradoxes.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Paradoxes
Normative model: Specifies how people should
behave
• People who behave according to axioms 1 – 7 should
make choices consistent with the maximization of
expected utility (normative model)
• It appears to be compelling, logical and fairly intuitive
• People do not necessarily make choices in accordance
with them.
• Research has established deviations from the normative
expected utility model. (descriptive model)
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Paradoxes
Deviations from utility theory
• “Paradoxes,” “irrational behavior”
Deviations can come from two sources:
• Violations of logic
• The procedure used to elicit these choices
We be must be careful to differentiate between those
sources of deviations.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Paradoxes
Framing effect: an individual’s risk attitude can
change depending on the way the decision problem is
posed (“framed”)
Violates procedure invariance
One of the most pervasive paradoxes in choice
behavior
Explicitly induced
A good example is presented in slide #3
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Paradoxes
Inconsistent choices from framing effects appear to
be that different points of reference are used to
frame the problem in two different ways.
• “Gains” and “losses” are treated differently
• Outcomes that are better than the reference point are defined
as “gains.”
• Outcomes that are worse than the reference point are defined
as “losses.”
• People tend to be risk-averse in dealing with gains but risk-
seeking in deciding about losses (prospect theory).
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Paradoxes
Reflection effect or loss aversion: asymmetry about
the reference point is reason gains and losses are
treated differently.
Losing $5 “hurts” more than finding $5 makes us
“feel good.”
Shown in the value function in the preceding slide
as the curve for losses being steeper than that of
gains
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Paradoxes
Allais Paradox:
A phenomenon in decision-making theory in which
people’s choices do not always align with expected
utility theory.
The Allais paradox shows that people’s choices do
not always follow this model, and that they may
make decisions that are not consistent with
maximizing their expected utility.
Expected utility theory is a mathematical model that predicts how people should make decisions in
order to maximize their expected utility (or satisfaction) from a given choice.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Paradoxes
Aspects of Allais Paradox: sources of inconsistencies
Certainty effect: individuals tend to place too much
weight on a certain outcome relative to uncertain
outcomes.
Sure-thing principle: preferences over lotteries or
risky prospects should depend only on the parts of
the lotteries that can lead to different outcomes
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Hedonic Framing
Hedonic framing or mental accounting: individuals
may implicitly frame decisions to enhance their appeal
due to the asymmetric shape of the value function
(see slide #19)
Violates expected utility theory
Two examples of hedonic framing/mental
accounting follow.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Example #1
“Who is happier?”
“A” who is given 2 lottery tickets that are both
winners, one for $25 and the other for $50
“B” who is given 1 lottery ticket that wins $75
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Example #1
Illustrates segregating gains:
Most people view A as being happier. We can
obviously see that both A and B are better off
by $75. Why does it matter the way in which
the $75 was received? The shape of the value
function indicates that when working with
gains, dividing a large gain into smaller ones
may be valued more by the receiver.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Example #2
“Who feels worse”?
“A” gets a letter from the Internal Revenue
Service saying he made a mistake on his
income tax return, and he owes $150.
“B” gets 2 letters, one saying he owes $100
on his Federal tax, and another saying he
owes $50 on his state income tax.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Example #2
Illustrates integrating losses:
Most consider B to feel worse, even though
they may realize they are both worse off by
$150.
People like to separate gains, but prefer to
combine losses. When the two are mixed, it
depends which is bigger.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Failure to Ignore Sunk Costs
Sunk costs
In accordance with utility theory, costs that cannot
be recouped should, rationally, be ignored in
decisions.
However, experimental evidence indicates that
many times sunk costs are not ignored and are
therefore incorrectly included in decision making.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Example
The effects of sunk costs on pizza consumption
Unknowing diners at a pizza restaurant were the
participants. Everyone paid $3 for an all-you-can-eat
lunch and then ordered from their server as many
pieces of pizza as they wanted. Half of the tables in
the restaurant were selected at random and given a
$3 refund before ordering. The other half did not get
the refund.
Who ate more pizza?
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Status Quo Bias
Status quo bias: a strong tendency to stick with the
status quo (current state of affairs) after a change in
the environment has occurred.
Sometimes this heuristic is rational.
• If nothing relevant has changed from situation 1 to
situation 2
• If making a change is costly, and furthermore is more
costly than any benefits derived from switching away
from the status quo
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Status Quo Bias
However research suggests that individuals often
have an irrational tendency to stay with the status
quo and can lead to poor decisions.
• For example, people will stay with the same bank even
though another bank is offering better interest rates,
cash for opening a new account, etc.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Implications
What are the implications for decision analysis of the
fact that people do not always operate in accordance
with the behavioral axioms?
1. How utility assessments should be made
2. How people actually make decision in the real-
world (e.g., workplace)
• Effects on company performance and on others who
depend on these decisions, e.g., managers and policy
makers
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Implications for Utility Assessment
Decision makers are expected to perform tasks
consistently with the axioms.
• Research shows they do not.
• This impacts the way utility functions are assessed.
In expected utility theory, increases in probability
are associated with constant change in value.
• People do not evaluate risky decisions this way.
• Small probabilities tend to be over-weighted; large
probabilities tend to be under-weighted.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Implications for Utility Assessment
Because of not following the behavioral axioms,
assessment techniques are distorted during decision
making.
When the consequences are gains, use of the CE
approach tends to result in more risk-averse
responses than does the PE approach.
When the consequences are losses, the CE
approach results in more risk-seeking behavior.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Managerial and Policy Implications
How do these inconsistencies in decision making
affect the “real-world”—government, businesses,
managers, policy-makers?
Fundamental issue: references points / status quo
• How are they determined or established?
• How are they perceived?
• Can they be changed or manipulated? Can the process
by which they are established be changed? Can their
perception be changed?
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Managerial and Policy Implications
Sunk Costs
Problem: Managers frequently ignore sunk costs
and remain committed to a project that obviously
has gone bad.
Solutions:
• Educate manager about the fallacy
• Help manager reframe the decision with a more
appropriate reference point
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Managerial and Policy Implications
Caveat: A problem’s frame may be specified with
equal validity in several ways.
For example, environmental protection
• People may view (i.e., frame) environmental programs
as anything from inconveniences (being forced to
separate recyclable materials from trash) to major
economic impediments (adhering to EPA’s complex
regulations).
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Managerial and Policy Implications
Sometimes we try too hard to eliminate uncertainty.
• Tendency to ignore uncertainty as if it isn’t there
• Costs to eliminate uncertainty may realistically be too
high – costs/benefits must be considered
Example – Medical treatment. “What is a life
worth?” We all would like to eliminate all risks of
someone dying unnecessarily. “Shouldn’t we spend
whatever it takes to be certain about diagnoses and
cures?” But is this realistic? Or even possible?
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Final Perspective
Do the inconsistencies from not following the
behavioral axioms invalidate expected utility theory?
Should think of decision analysis as a tool that
helps us…
• Make better decisions
• Structure and model problems
• Consider the process consciously
All of these procedures, and more, in decision
analysis can attenuate—even eliminate—biases.
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.
Summary
In this chapter, you have learned:
About the foundation and axioms that underlie the
expected utility theory
That if we are consistent with this foundation and the
behavior patterns that we call axioms, we will maximize
utility
That, while we agree with the above, we are often
inconsistent with the foundation and the axioms
That many heuristics and biases have been identified,
which is a first step toward controlling their harmful
influence
© 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a password-
protected website for classroom use.