Distortions In Deriving
Preferences
Changes matter more than states. People’s
preferences are especially sensitive to changes.
Suppose you are asked two questions:
A:
Imagine then you are richer by Euro 20,000
than you are today. Would you prefer an additional
gain of 5,000 for sure or a 50–50 chance for a gain
of 10,000 or nothing? Make one choice.
B:
Imagine then you are richer by Euro 30,000
than you are today. Would you prefer an additional
loss of 5,000 for sure or a 50–50 chance for a loss
3.11
of 10,000 or nothing? Make one choice.
Lecture
7
Wednesday, 16 March 2016
12:09 AM
Distortions In Deriving
Preferences
Although the final outcomes in the two problems are
exactly the same (€25,000 or 50/50 €20,000 or
€30,000), most people choose the gamble in Question
A and the sure loss in Question B. Apparently, they
tend to favour the narrow framing based on gains and
losses rather than the broader (and more relevant)
framing based on the final wealth.
For a purely mathematical view see; can a gamble
ever be right or wrong? Is there an objective way to
say whether a particular bet was good value or not?
3.22
Distortions In Deriving
Preferences - Loss Aversion
People’s sensitivity to losses is higher than their
sensitivity to gains. Suppose you are asked the
question:
Consider a bet on the toss of the coin. If heads, you
lose Euro 100. What is the minimum gain, if tails, that
would make you accept the gamble? Make a choice.
Most answers typically fall in the range from 200 to
250, which reflects a sharp asymmetry between the
values that people attach to gains and loses.
3.33
Distortions In Deriving
Preferences - Loss Aversion
One ubiquitous pattern stands out: Losses resonate
more than gains.
In a wide variety of domains, people are more averse
to losses than they are attracted to same-sized
gains.
One of the main realms where loss aversion plays out
is in preferences over wealth levels.
3.44
Distortions In Deriving
Preferences - Loss Aversion
What distinguishes loss aversion from conventional
risk aversion is that people are significantly “risk
averse” for even small amounts of money.
People dislike losing $10 more than they like gaining
$11, and hence prefer their status quo to a 50/50
bet of losing $10 or gaining $11.
Tversky and Kahneman (1991) suggest that in most
domains where sizes of losses and gains can be
measured, people value moderate losses roughly
twice as much as equal-sized gains.
3.55
Distortions In Deriving
Preferences - Loss Aversion
Finally, in prospect theory (which is a key concept in
behavioural economics), the pain associated with a
possible loss is much greater than the pleasure
associated with a gain of the same magnitude. In the
strip, Dilbert's garbage man clearly understands this
concept much better than Dilbert. Cartoon (Kramer
2014)
Other models expressing loss aversion with a linear
utility above the target and a specific concave utility
below the target have been suggested (for a short
and useful review see Jarrow and Zhao, 2006).
3.66
Distortions In Deriving
Preferences - Loss Aversion
People fear losing more than they desire to win. This
phenomenon, first demonstrated by Tversky and
Kahneman (1991), is known as loss aversion, and it shows up
everywhere.
Kahneman (2011) describes how economists Pope and
Schweitzer (2011) reasoned that golf provides a perfect
example of a reference point: par.
3.77
Distortions In Deriving
Preferences - Loss Aversion
Every hole on the golf course has a number of strokes
associated with it; the par number provides the baseline
for good — but not outstanding — performance. For a
professional golfer, a birdie (one stroke under par) is a
gain, and a bogey (one stroke over par) is a loss. The
economists compared two situations a player might face
when near the hole:
putt to avoid a bogey
putt to achieve a birdie
3.88
Distortions In Deriving
Preferences - Loss Aversion
Every stroke counts in golf, and in professional golf every
stroke counts a lot. According to prospect theory,
however, some strokes count more than others. Failing to
make par is a loss, but missing a birdie putt is a foregone
gain, not a loss. Pope and Schweitzer reasoned from loss
aversion that players would try a little harder when
putting for par (to avoid a bogey) than when putting for a
birdie. They analysed more than 2.5 million putts in
exquisite detail to test that prediction.
3.99
Distortions In Deriving
Preferences - Loss Aversion
They were right. Whether the putt was easy or hard, at
every distance from the hole, the players were more
successful when putting for par than for a birdie. The
difference in their rate of success when going for par (to
avoid a bogey) or for a birdie was 3.6%. This difference is
not trivial. Tiger Woods was one of the “participants” in
their study. If, in his best years, Tiger Woods had
managed to putt as well for birdies as he did for par, his
average tournament score would have improved by one
stroke and his earnings by almost $1 million per season.
3.10
10
Distortions In Deriving
Preferences - Loss Aversion
These fierce competitors certainly do not make a
conscious decision to slack off on birdie putts, but their
intense aversion to a bogey apparently contributes to
extra concentration on the task at hand.
The study of putts illustrates the power of a theoretical
concept as an aid to thinking. Who would have thought it
worthwhile to spend months analysing putts for par and
birdie? The idea of loss aversion, which surprises no one
except perhaps some economists, generated a precise and
non-intuitive hypothesis and led researchers to a finding
that surprised everyone — including professional golfers.
“Golf is a good walk spoiled” Mark Twain
3.11
11
(30 November 1835 Florida - Missouri)
Loss Aversion - Barings Bank
How destructive is loss aversion? Perhaps the most familiar
case of loss aversion or “get-evenitis,” occurred in 1995,
when 28-year-old Nicholas Leeson caused the collapse of
his famous employer, the 233-year-old Barings Bank. At the
end of 1992, Leeson had lost about £2 million, which he hid
in a secret account. By the end of 1993, his losses were
about £23 million, and they mushroomed to £208 million at
the end of 1994 (at the time, this was $512 million).
Instead of admitting to these losses, Leeson gambled more
of the bank’s money in an attempt to “double-up and catchup.”
Rogue Trader, Pub. Pathé, 1999 here also Rogue Trader: How I Brought
Down Barings Bank and Shook the Financial World, Nick Leeson, Pub.
3.12
12
Little, Brown & Company, 1996 here
Loss Aversion - Barings Bank
On February 23, 1995, Leeson’s losses were about £827
million ($1.3 billion) and his trading irregularities were
uncovered. Although he attempted to flee from
prosecution, he was caught, arrested, tried, convicted, and
imprisoned. Also, his wife divorced him (Jordan et al. 2012).
For a review that focuses on banks rather than borrowers
and associated risks see Beatty and Liao (2014). There is a
very large literature on bank debt contracting, the
literature is surveyed by Armstrong et al. (2010).
3.13
13
Distortions In Deriving
Preferences - Risk Aversion
Women are found to be more risk averse in making
financial decisions than men (Donkers and Van Soest,
1999; Powell and Ansic, 1997; Weber et al., 2002).
Women also tend to own less risky assets than single men
or married couples and reduce their risky assets when the
number of children increases, in contrast to single men
and married couples (Jianakoplos and Bernasek, 1998).
Furthermore, older people tend to take less financial
risks than younger people (Jianakoplos and Bernasek,
2006).
3.14
14
Distortions In Deriving
Preferences - Risk Aversion
On the other hand, it has been shown that it was possible
to teach effective risk diversification (Hedesström et al.,
2006). “In 4 experiments, undergraduates made
hypothetical investment choices. … In order to counteract
naïve diversification, novice investors need to be better
informed about the rationale underlying recommendations
to diversify.”
3.15
15
Distortions In Deriving
Preferences - Risk Aversion
People’s financial history has a strong impact on their
taste for risk. Malmendier and Nagel (2011) investigated
whether individual experiences of macroeconomic shocks
affect financial risk taking. As has often been suggested
for the generation that experienced the Great
Depression (1930s). Using data from the Survey of
Consumer Finances from 1960 to 2007. They found that
individuals who have experienced low stock market
returns throughout their lives so far report lower
willingness to take financial risk. They are less likely to
participate in the stock market. Invest a lower fraction
of their liquid assets in shares if they participate. They
are more pessimistic about future share returns.
3.16
16
Distortions In Deriving
Preferences - Risk Aversion
Exposure to economic turmoil appears to dampen people’s
appetite for risk irrespective of their personal financial
losses. That is the conclusion of Knüpfer et al. (2013) who
suggest that labour market experiences are a natural
candidate for explaining portfolio heterogeneity. In the
early 1990s a severe recession caused Finland’s GDP to
sink by 10% and unemployment to soar from 3% to 16%.
Using detailed data on tax, unemployment and military
conscription, the authors were able to analyse the
investment choices of those affected by Finland’s “Great
Depression”.
3.17
17
Distortions In Deriving
Preferences - Risk Aversion
The results suggest that workers who have experienced
more adverse labour market conditions are significantly
less likely to invest in risky assets. The decrease in risky
investment is robust to controls for parental variables,
family fixed effects, and cognitive ability. It is not fully
attributable to the impact labour market shocks have on
future income, unemployment risk, and wealth
accumulation. They found that those hit harder by
unemployment were less likely to own stocks a decade
later. Individuals’ personal misfortunes, could explain at
most half of the variation in stock ownership. They
attribute the remainder to “changes in beliefs and
preferences” that are not easily measured.
3.18
18
Distortions In Deriving
Preferences - Risk Aversion
Financial trauma appears to dampen people’s appetite for
risk. Guiso et al. (2013) examined the investments of
several hundred clients of a large Italian bank in 2007
and again in 2009 (i.e. before and after the plunge in
global stock markets). The authors also asked the clients
about their attitudes towards risk and got them to play a
game modelled on a television show in which they could
either pocket a small but guaranteed prize or gamble on
winning a bigger one.
3.19
19
Distortions In Deriving
Preferences - Risk Aversion
Risk aversion, by these measures, rose sharply after the
crash, even among investors who had suffered no losses in
the stock market. They found that both a qualitative and
a quantitative measure of risk aversion increases
substantially after the crisis. After considering standard
explanations, they investigated whether this increase
might be an emotional response (fear) triggered by a
scary experience. The reaction to the financial crisis, the
authors conclude, looked less like a proportionate
response to the losses suffered and “more like oldfashioned ‘panic’.” To show the plausibility of this
conjecture, they conducted a laboratory experiment.
3.20
20
Distortions In Deriving
Preferences - Risk Aversion
The authors’ conclusions were reinforced by a separate
test administered to a few hundred university students.
About half were asked to watch a five-minute excerpt of
a gruesome torture scene from a horror film. Then, the
entire group answered the same questions about risk as
the Italian bank’s clients. Watching the horror movie
increased the students’ aversion to risk by roughly as
much as the financial crisis had chastened the bank’s
clients, although not among those who claimed to like
horror movies. They found that subjects who watched a
horror movie have a certainty equivalent that is 27%
lower than the ones who did not, supporting the fearbased explanation.
3.21
21
Risk Aversion – Seasonality
Does the Caveman Within Tell You How to Invest? Changes in human biology across the
seasons helps shape investment decisions - Psychology Today - 18 August 2014 - Lisa
Kramer
Is there a rational theory of human behaviour that helps
explain both the “Sell in May, then go away” effect
observed in risky stock markets and an opposing seasonal
cycle observed in safe Treasury bond markets. The paper
by Kamstra et al. (2014 KKLW) is a bit technical, but it
can be summed up easily with reference to our caveman
ancestors. The study finds that the historical patterns of
stock and bond returns through the seasons implies not
only seasonally changing investor risk aversion, but also
seasonal changes in the way investors decide between
consuming now versus saving for the future.
3.22
22
Risk Aversion – Seasonality
Many people experience severe seasonal depression
(seasonal affective disorder - SAD), during the dark
seasons of autumn and winter. When people get
depressed in the winter, they become more averse to
financial risk. The implication is a much larger reward, on
average, for investors who are willing to hold risky stock
during the dark seasons. Once daylight returns, investors
become much more tolerant of risk, and with droves of
people heading back into stocks in the spring, the rate of
return investors earn holding stocks reduces, on average,
leading to lower rewards for holding risk in the spring and
summer. Lo and behold, Wall Street has an adage for
that: “Sell in May, and go away” (KKLW).
3.23
23
Risk Aversion – Seasonality
How does an examination of our ancestors help shed light
on all of this? The likelihood of our ancestors surviving
another year was almost certainly enhanced by certain
traits, such as willingness to save in the spring and
summer so one could consume from stores of food in the
autumn and winter. If our ancestors had acted like the
proverbial grasshopper during the spring and summer —
wildly consuming instead of judiciously saving — the odds
of surviving through the next autumn and winter were
certainly worsened. Thus, saving instead of consuming
during the spring and summer would be the preferred
behaviour for folks aiming to maximize the odds of
survival. Additionally, by adopting cautious, “risk averse”
habits through the autumn and winter, hunkering down
until the daylight rebounded, they would increase their
3.24
24
odds of surviving through to the spring (KKLW).
Risk Aversion – Seasonality
What are the key elements? Be aware of the way the
changing seasons can influence your mood and ultimately
your behaviour. Be cautious about making important
financial decisions during particularly emotional times.
Recognise that what might look like a great investment
idea in the summer might not feel so appropriate once the
winter doldrums set in. Try to build a portfolio that can
endure the seasons. Investors who “buy and hold” tend to
do better on average than those who trade frequently in
an attempt to outperform the market. Accordingly,
consider holding investments you’ll be comfortable holding
through thick and thin.
Kamstra, M.J., Kramer, L.A., Levi, M.D. and Wang T. 2014 “Seasonally Varying
Preferences: Theoretical Foundations for an Empirical Regularity” The Review of Asset
3.25
25
Pricing Studies 4(1) 39-77.
Risk Aversion – Seasonality
Daylight-Saving Time Changes, Anxiety, & Investing. Disruptions in
your sleep can have an impact on your portfolio - Psychology Today - 30
October 2013 - Lisa Kramer
During the first weekend of November, people will turn
their clocks back an hour. The gain or loss of an hour on
the clock typically translates immediately into the gain or
loss of an hour of sleep. Psychologists refer to the fallout
from such disruptions as “sleep desynchronosis”, with the
symptoms of the changed sleep habits closely resembling
jet lag. And the phenomenon can have significant
consequences. For instance, psychologists have noticed that
whenever we shift the clocks to or from daylight-saving
time, car accident rates tend to rise, likely due to the
3.26
26
cognitive changes that accompany disrupted sleep habits.
Risk Aversion – Seasonality
Perhaps surprisingly, the phenomenon is observed
whether people are gaining or losing an hour of sleep.
Several large-scale disasters have been associated with
lack of sleep or disrupted sleep due to shift work,
including the Exxon Valdez oil spill, the space shuttle
Challenger explosion, the Three Mile Island nearmeltdown, and the Chernobyl nuclear accident. Adverse
effects of daylight-saving time changes can also spill over
into financial markets, seemingly through increased
anxiety that accompanies altered sleep patterns.
Daylight-Saving Time Changes, Anxiety, & Investing |
Psychology Today
3.27
27
Risk Aversion – Seasonality
In the language of financial economists, an increase in
anxiety translates into greater reluctance to bear
financial risk (or greater risk aversion). If investors wake
up on the Monday following the time change feeling more
anxious (and more risk averse) than usual, they may be
less willing to buy risky stock and could even consider
selling the stocks they already own. With many investors
simultaneously experiencing such a shift in their
sentiment , the result is often a drop in stock markets on
the Monday following the time change. Of course financial
markets are impacted by many different factors (not
least important of which is fundamental economic news),
so naturally the stock market could go up or down
following any given daylight-saving time change (Kamstra
28
et al. 2000, Pinegar 2002 and Kamstra et al. 2002). 3.28
Risk Aversion – Seasonality
The study (references follow) considers stock market
returns from four countries, some of which change the
clocks on different dates. They found the market
downturn associated with daylight-saving time changes
amounted to a single-day loss of $31 billion in US
markets, on average.
3.29
29
Risk Aversion – Seasonality
The finding does not imply ordinary investors should do
anything drastic in preparation for the time change. I am
definitely not counselling anyone to time their purchase or
sale of securities in accordance with the time change. In
fact, experience tells us the best way to navigate through
our emotions in the context of investing is to avoid making
important decisions whenever feelings come in to play.
Making investing decisions during emotional times often
results in suffering dramatic financial consequences. Think
of all the people who panicked during the recent financial
crisis, selling just when their fears – and markets – were at
their worst. Many of them ended up liquidating their assets
at steeply discounted prices. Had they just sat tight and
stuck to a buy and hold approach, they would eventually
3.30
30
have noticed their stocks resuming their previous values.
Risk Aversion – Seasonality
At the best of times, people can be forgiven for feeling a
sense of anxiety when it comes to investing. Emotions can
ride even higher than usual under some conditions,
including times of sleep disruption such as those
associated with daylight-saving time changes. By
remaining calm and avoiding impulsive, emotion-driven
investment decisions, there is no need to lose sleep over
the market.
Kamstra, M.J., Kramer, L.A. and Levi, M.D. 2000 “Losing sleep at the market:
The daylight saving anomaly” American Economic Review 90(4) 1005-1011.
Pinegar, J.M. 2002 “Losing sleep at the market: Comment” American
Economic Review 92(4) 1251-1256.
Kamstra, M.J., Kramer, L.A. and Levi, M.D. 2002 “Losing sleep at the market:
The daylight saving anomaly: Reply” American Economic Review 92(4) 12571263.
3.31
31
Or Is It A Gut Reaction?
In new research (Thaiss et al. 2014), the microbes in the faeces of humans
and mice were analysed, and it was discovered that gut microbes follow a
rhythmic pattern throughout the day. The cycle depends on eating habits and
the circadian cycle of the human or mouse.
The microbes were disrupted when the mice were exposed to an abnormal
eating schedule and changes in their exposure to light and dark, the study
found. In two people who suffered from jet lag, certain types of bacteria
became more common. The germs are linked to obesity and problems in the
body's metabolic system, according to the researchers.
Jet lag can cause obesity by disrupting the daily rhythms of gut microbes ScienceDaily - 16/Oct/2014
Jet lag doesn’t just leave travellers feeling rotten – it is also being blamed
for making them fatter - Telegraph - 16/Oct/2014
3.32
32
Or Is It A Gut Reaction?
Thaiss et al. “Transkingdom Control of Microbiota Diurnal Oscillations
Promotes Metabolic Homeostasis”, Cell, 2014.
3.33
33
Risk Aversion – A Counter
Previous studies of loss aversion in decisions under risk
have led to mixed results. Losses appear to loom larger
than gains in some settings, but not in others. A paper by
Ert and Erev (2013) highlighted six experimental
manipulations that tend to increase the likelihood of the
behaviour predicted by loss aversion.
3.34
34
Risk Aversion – A Counter
These manipulations include:
1
2
3
4
5
6
framing of the safe alternative as the status quo;
ensuring that the choice pattern predicted by loss
aversion maximizes the probability of positive
(rather than zero or negative) outcomes;
the use of high nominal (numerical) payoffs;
the use of high stakes;
the inclusion of highly attractive risky prospects
that creates a contrast effect;
the use of long experiments in which no feedback is
provided and in which the computation of the
3.35
35
expected values is difficult.
Risk Aversion – A Counter
Their results suggest the possibility of learning in the
absence of feedback: The tendency to select simple
strategies, like “maximize the worst outcome” which
implies “loss aversion”, increases when this behaviour is
not costly. Theoretical and practical implications are
discussed (Ert and Erev, 2013).
3.36
36
Distortions In Deriving Preferences
- Endowment Effect
Loss aversion is related to the striking endowment
effect identified by Thaler (1980, 1985).
Once a person comes to possess a good, they
immediately value it more than before they
possessed it.
How would you feel if you lost everything you owned,
even if you were financially compensated? Like part
of you had died? Or liberated? Examine the
psychology of our lifelong relationship with objects
(Jarrett, 2013).
3.37
37
Distortions In Deriving Preferences
- Endowment Effect
Kahneman et al. (1990) tested the endowment effect
in a series of experiments, conducted in a classroom
setting.
In one of these experiments a decorated mug (retail
value of about $5) was placed in front of one third of
the seats after students had chosen their places.
3.38
38
Distortions In Deriving Preferences
- Endowment Effect
All participants received a questionnaire.
The form given to the recipients of a mug (the
“sellers”) indicated “You now own the object in your
possession. You have the option of selling it if a
price, which will be determined later, is acceptable
to you. For each of the possible prices below indicate
whether you wish to
(x)
Sell your object and receive this price;
(y)
Keep your object and take it home with you.”
The subjects indicated their decision for prices
ranging from $0.50 to $9.50 in steps of 50 cents.
3.39
39
Distortions In Deriving Preferences
- Endowment Effect
Some of the students who had not received a mug
(the “choosers”) were given a similar questionnaire,
informing them that they would have the option of
receiving either a mug or a sum of money to be
determined later.
They indicated their preference between a mug and
sums of money ranging from $0.50 to $9.50.
The choosers and the sellers face precisely the same
decision problem, but their reference states differ.
3.40
40
Distortions In Deriving Preferences
- Endowment Effect
The choosers face a positive choice between two
options that dominate (their reference state).
The sellers must choose between retaining the
status quo (the mug) or giving up the mug in exchange
for money.
Thus, the mug is evaluated as a gain by the choosers,
and as a loss by the sellers.
Loss aversion entails that the rate of exchange of
the mug against money will be different in the two
cases.
3.41
41
Distortions In Deriving Preferences
- Endowment Effect
Indeed, the median value of the mug was $7.12 for
the sellers and $3.12 for the choosers in one
experiment, $7.00 and $3.50 in another. The
difference between these values reflects an
endowment effect, which is produced, apparently
instantaneously, by giving an individual property
rights over consumption goods.
3.42
42
Distortions In Deriving Preferences
- Endowment Effect
The behaviour described here is usefully
conceptualised as a case of loss aversion comparable
to that identified in choice among lotteries.
Individuals who are randomly given mugs treat the
mug as part of their reference levels or endowments,
and consider not having a mug to be a loss, whereas
individuals without mugs consider not having a mug as
remaining at their reference point.
3.43
43
Distortions In Deriving
Preferences - Status Quo Bias
As established by Knetsch and Sinden (1984),
Samuelson and Zeckhauser (1988), and Knetsch
(1989), a comparable phenomenon - the status quo
bias - holds in goods choice problems.
Here, loss aversion implies that an individual's
willingness to trade one object for another depends
on which object they begin with: Individuals tend to
prefer the status quo to changes that involve losses
in some dimensions, even when these losses are
coupled with gains in other dimensions.
3.44
44
Distortions In Deriving
Preferences - Status Quo Bias
Knetsch and Sinden (1984) and Knetsch (1989), for
instance, demonstrated the status quo bias by
randomly giving one set of students candy bars, and
the remaining students decorated mugs.
Later, each student was offered the opportunity to
exchange their gift for the other one - a mug for a
candy bar or vice versa.
90% of both mug-owners and candy-owners chose not
to trade.
3.45
45
Distortions In Deriving
Preferences - Status Quo Bias
Because the goods were allocated randomly and
transaction costs were minimal, the different
behaviour for the two groups of subjects must have
reflected preferences that were induced by the
allocation.
3.46
46
Distortions In Deriving
Preferences - Status Quo Bias
Other experiments have shown that the more
choices you are given, the more pull the status quo
has. More people will, for instance, choose the status
quo when there are two alternatives to it rather than
one: For example A and B instead of just A.
Why? Choosing between A and B requires additional
effort; selecting the status quo avoids that effort.
In business, where sins of commission (doing
something) tend to be punished much more severely
than sins of omission (doing nothing), the status quo
3.47
47
holds a particularly strong attraction.
Distortions In Deriving
Preferences - Choice
Lipowski (1970) call this problem an approach-approach
conflict: faced with enticing options, you find yourself
unable to commit to any of them quickly. And even when
you do choose, you remain anxious about the opportunities
that you may have lost: maybe the “grass is greener”. He
maintains that it is specifically the overabundance of
attractive alternatives, aided and abetted by an affluent
and increasingly complex society that leads to conflict,
frustration, unrelieved appetitive tension, more approach
tendencies and more conflict — a veritable vicious cycle.
That cycle, in turn, likely has far-reaching and probably
harmful effects on the mental and physical health of
3.48
48
affected individuals.
Distortions In Deriving
Preferences - Choice
Iyengar and Lepper (2000) revived the idea of conflict created by an
overabundance of choice — a concept called the paradox of choice
(Schwartz 2004) and focusing on the concept of cognitive demands.
When shoppers had to choose between jams or chocolates, it was
found, they were more likely to make a selection when faced with six
choices than when presented with twenty-four or thirty. They were
also more satisfied with their ultimate selection. Too much choice
would reduce motivation. That could be because an abundance of
options may simultaneously attract and repel choice-maker, an
emotion-based explanation. In a series of imaging studies Shenhav
and Buckner (2014) observed students making various choices when
inside an fMRI scanner. They found that given more “good” choices,
makes you feel more anxious.
3.49
49
Status Quo Bias - Personal C
Current Bank Account
A related study by the Office of Fair Trading (2008)
examined the psychology of personal current bank
account usage, with a focus upon account switching
and bank charges.
Each year, only a small fraction of people switch
current accounts and the overall switching rate is
low. Each year, a large proportion of people incur
bank charges.
Six psychological factors that may affect switching
and charge-incurring behaviour were considered.
3.50
50
Status Quo Bias - Personal C
Current Bank Account Bias
Key findings include:
1.
Interviewees believed that, when switching,
payments may be missed and were averse to possible
practical losses. The suggestion is that people are
not concerned about general financial loss during
switching, but more concerned about the
inconvenience and hassle of losing features or having
to correct missed payments.
3.51
51
Status Quo Bias - Personal C
Current Bank Account
Key findings include:
2.
However interviewees' perceptions of their
control over switching were high, suggesting that,
despite problems, they believed that they could
switch as long as they had the necessary resources
and ability to change and were not limited by outside
forces.
3.52
52
Status Quo Bias - Personal C
Current Bank Account
Key findings include:
3.
Interviewees showed a strong focus on the
here-and-now at the expense of a future orientation,
and may constantly defer taking action to switch. In
order for a person to take action, the expected
future gain received will need to be financially much
larger than the future consequence.
3.53
53
Status Quo Bias - Personal C
Current Bank Account
Key findings include:
4.
Interviewees were overconfident in their
financial management, and underestimated the
likelihood that they would become overdrawn and be
charged. This overconfidence means that people
probably underestimate the cost of banking and are
more optimistic about the cost of banking. This
optimism could result from psychological
overconfidence in one's own abilities, but it may also
result from a failure to correctly identify
unpredictable outside consequences that may cause
3.54
54
one to become overdrawn.
Status Quo Bias - Personal C
Current Bank Account
Key findings include:
5.
Interviewees reported spending little time
thinking about their finances.
3.55
55
Status Quo Bias - Personal C
Current Bank Account
Key findings include:
6.
Perceptions of charges showed that, in
principle, the existence of charges were not viewed
as particularly unfair, with more favourable
perceptions associated with increased awareness and
advance warning of charges.
3.56
56
Status Quo Bias - Personal C
Current Bank Account
The UK's 46 million current account holders will be able to
switch banks in seven days from next month.
After two years of preparation, the Payments Council has
confirmed a new switching service will start on 16 September
2013. (The Payments Council was an organisation of financial
institutions in the United Kingdom, that set strategy for UK
payment mechanisms from 2007 until 2015.)
Until now, transferring an account to a new provider has taken
up to 30 days.
In anticipation of the new scheme, two banks are already
offering customers an incentive of up to £125 to switch their
current accounts to them.
3.57
57
Status Quo Bias - Personal C
Current Bank Account
Watchdog tells banks to work harder for customers - FT - 22 Oct 2015
Banks in the UK will be forced to guide customers to
rival current accounts in an attempt to stoke
competition, but will not have to abandon free banking
for people in credit or face being broken up.
The competition watchdog said on Thursday that the
largest lenders — Lloyds Banking Group, Barclays, Royal
Bank of Scotland and HSBC — must alert customers to
switch to another current account provider in situations
such as a branch closure or technology glitches.
3.58
58
Status Quo Bias - Personal C
Current Bank Account
The account switching service is being launched
following a recommendation from the Independent
Commission on Banking two years ago, which said that
people only changed bank accounts once every 26
years on average.
BBC News - Bank account switching service to launch in September
16 August 2013
3.59
59
Status Quo Bias - Personal C
Current Bank Account
Customers reluctant to ditch big four UK banks - FT - 3 Oct 2015
About 16 per cent would likely switch to a so-called
challenger bank, such as Metro or TSB, within the
next two years, up from 13 per cent last year,
according to research by law firm Pinsent Masons
and YouGov.
But three-quarters still expect to be banking with a
traditional high street bank in two years’ time,
undermining attempts to inject competition in the
sector.
3.60
60
Status Quo Bias - Avoidance
Hammond et al., 2006
1. Always remind yourself of your objectives and
examine how they would be served by the status
quo. You may find that elements of the current
situation act as barriers to your goals.
2. Never think of the status quo as your only
alternative. Identify other options and use them
as counterbalances, carefully evaluating all the
pluses and minuses.
3. Ask yourself whether you would choose the
status-quo alternative if, in fact, it weren't the
status quo.
3.61
61
Status Quo Bias - Avoidance
4. Avoid exaggerating the effort or cost involved in
switching from the status quo.
5. Remember that the desirability of the status quo
will change over time. When comparing
alternatives, always evaluate them in terms of the
future as well as the present.
6. If you have several alternatives that are superior
to the status quo, don't default to the status quo
just because you're having a hard time picking the
best alternative. Force yourself to choose.
3.62
62
Status Quo Bias – Luke’s
Axiom Of Choice
Luce’s choice axiom (Luce 1959 and 1977) is a theory of individual
choice behaviour that has proven to be a powerful tool in the
behavioural sciences for over 50 years. Luce’s choice axiom is
grounded in two fundamental properties: choice is probabilistic and
the probability of choosing an option from one set of alternatives is
related to the probability of choosing the same option from a
different set.
The probability of selecting one item over another from a pool of
many items is not affected by the presence or absence of other
items in the pool. Selection of this kind is said to have
“independence from irrelevant alternatives”.
For instance do you usually choose the same meat (chicken chow
3.63
63
mein or tandoori chicken) from the menu?
Status Quo Bias – In Marketing
Luce (1998) conducted a series of experiments which
created purchasing decisions that pitted important
values against each other. The more vivid she made
these decisions for her subjects, the more intense
the negative emotions that were aroused. Luce
discovered that as these negative emotions became
more intense, subjects' choice of the status quo
alternative became more common.
3.64
64
Status Quo Bias – In Marketing
Everyone in sales and marketing is familiar with the
status quo alternative. That's the choice customers
make when they decide not to purchase anything at
all and “make do” with what they have, e.g., get a few
more miles out of the old Chevy or wear last year's
bathing suit one more year. The attractiveness and
popularity of the status quo alternative is a curse
which bedevils salespeople, and we call it the status
quo bias.
3.65
65
Status Quo Bias – In Marketing
Salespeople attribute many causes for the status quo
bias, but Luce's experiments showed that people
select the status quo primarily to relieve negative
emotions - emotions aroused by competing values in
the decision. These experiments also showed that
negative emotions can be lessened if people are
reminded that a status quo alternative is available.
Ordinarily, the only time salespeople call attention to
the status quo alternative is to subtly disparage it,
thereby hoping to discourage its selection, but
Luce's experiments suggest this is a mistake.
3.66
66
Status Quo Bias – In Marketing
Luce points out that there are actually three status
quo alternatives in most choice situations:
1) “make do” with what you have,
2) continue to search, and
3) select the dominant alternative.
The key here is in that third choice.
3.67
67
Status Quo Bias – In Marketing
Consider, for example, a car salesperson who
recognises negative emotions building in a customer
who is trying to resolve the safety versus economy
conflict. This salesperson could call attention to the
status quo alternative of the dominant choice by
saying: “Most people resolve this problem by
selecting model ABC. It offers the most popular
balance of economy and safety.”
3.68
68
Status Quo Bias – In Marketing
The immediate effect of this tactic should be to
reduce the negative emotions customers are
experiencing, and Luce's research also reveals that it
makes this choice more common.
Salespeople lose credibility when they use their
knowledge of psychology to manipulate and exploit
customers, but they gain credibility when they use
this same knowledge to help customers make
difficult decisions. Calling attention to all three
status quo alternatives seems to fall into this latter
category.
3.69
69
Status Quo Bias – Utility
Function
Knetsch (1989) experimentally demonstrates that such
preferences can be usefully captured by utility functions
defined over reference levels as well as consumption levels.
The most prominent result is loss aversion - the observation
that a loss is given greater value than a gain of an equal size
- resulting in the S-shaped utility function (Dacey 2003),
for reviews see Camerer (1995) and Rabin (1998).
In reality what is utility? If you have a bad back, would you
bend down to pick up a penny? What about £50!
Consider the following rather stylised graph.
3.70
70
Status Quo Bias – Utility
Function
The loci of points, IA and IB, represent indifference curves
for bundles of Goods 1 and 2 (say apples and pears) for a
consumer at, respectively, reference point A and reference
point B.
3.71
71
Status Quo Bias – Utility
Function
Such indifference curves capture the status quo bias,
because they imply that the consumer strictly prefers A
over B if they are at A.
3.72
72
Status Quo Bias – Utility
Function
Such indifference curves capture the status quo bias,
because they imply that the consumer strictly prefers B
over A if they are at B.
3.73
73
Status Quo Bias – Utility
Function
More generally, they will tend to prefer bundles that
avoid losses of any goods; thus, they will prefer X over Y
if they are at A.
3.74
74
Status Quo Bias – Utility
Function
More generally, they will tend to prefer bundles that
avoid losses of any goods; thus, they will prefer Y over X
if they are at B.
3.75
75
Distortions In Deriving Preferences
- Probability Weighting
With respect to the expected utility paradigm,
people tend to weigh probabilities differently.
Suppose you are asked:
3.76
76
Distortions In Deriving Preferences
- Probability Weighting
Given a chance for a gain of Euro 20,000, would you pay
more to raise the probability of a gain from 0% to 1%,
from 41% to 42%, or from 99% to 100%? Your choice?
While expected utility predicts that the answer should be
the same, most people would pay significantly less for
raising the probability to 42%.
In particular, low probabilities are over weighted: people
tend to find a 1% chance of winning Euro 1,000 preferable
to a sure Euro 10.
3.77
77
Distortions In Deriving Preferences
- Probability Weighting
Empirical studies have shown that decision makers do
not usually treat probabilities linearly. Instead,
people tend to overweight small probabilities and
underweight large probabilities. One way to model
such distortions in decision making under risk is
through a probability weighting function (Gonzalez
and Wu 1999, Prelec 1998, Rachlin et al. 1991,
Stewart et al. 2014).
3.78
78
Distortions In Deriving Preferences
- Diminishing Sensitivity
In addition to loss aversion, another important
feature of how people assess departures from their
reference levels is that they have diminishing
sensitivity - the marginal change in perceived well
being is greater for changes that are close to one’s
reference level than for changes that are further
away.
3.79
79
Distortions In Deriving Preferences
- Diminishing Sensitivity
As with loss aversion, Kahneman and Tversky (1979,
p. 278) argue that diminishing sensitivity reflects a
more fundamental feature of human cognition and
motivation: Many sensory and perceptual dimensions
share the property that the psychological response is
a concave function of the magnitude of physical
change (see definition and later slide) .
Definitions Slide
3.80
80
Distortions In Deriving Preferences
- Diminishing Sensitivity
For example, it is easier to discriminate between a
change of 3º and a change of 6º in room
temperature, than it is to discriminate between a
change of 13º and a change of 16º.
We propose that this principle applies in particular to
the evaluation of monetary changes.
Thus, the difference between a gain of 100 and a
gain of 200 appears to be greater than the
difference between a gain of 1,100 and a gain of
1,200.
3.81
81
Distortions In Deriving Preferences
- Diminishing Sensitivity
Similarly, the difference between a loss of 100 and a
loss of 200 appears greater than the difference
between a loss of 1,100 and a loss of 1,200, unless
the larger loss is intolerable.
Thus, we hypothesize that the value function for
changes of wealth is normally concave above the
reference point and often convex below it. That is,
the marginal value of both gains and losses generally
decreases with their magnitude.
3.82
82
Distortions In Deriving Preferences
- Diminishing Sensitivity
Thus, we hypothesize that the value function for changes
of wealth is normally concave above the reference point and
often convex below it. That is, the marginal value of both
gains and losses generally decreases with their magnitude.
For those interested in evaluating
this function for “real” individuals,
refer to Swalm (1966) Booij and van
de Kuilen (2009).
It is suggested that risk preference
is related to skewness (the third
standardised moment) and aversion
to kurtosis (the fourth standardised
moment) (Ebert, 2013).
3.83
83
Distortions In Deriving Preferences
- Diminishing Sensitivity
It is suggested that risk preference is related to skewness (the third
standardised moment) and aversion to kurtosis (the fourth
standardised moment) (Ebert, 2013). Hedge funds returns often
exhibit high negative skewness and positive excess kurtosis
(Zakamouline and Koekebakker, 2009 ).
3.84
84
Distortions In Deriving Preferences
- Diminishing Sensitivity
Diminishing marginal utility
Increasing marginal utility
Constant marginal utility of
of wealth
of wealth
wealth
EU of the gamble < utility
EU of the gamble > utility
EU of the gamble = utility
of a certain wealth
of a certain wealth
of a certain wealth
Prefer certain income than
Prefer
uncertain
than certain income with
certain
the same expected value
uncertain income with the
(risky)
income
with the same expected
uncertain
income
value
Indifference
between
income
than
same expected value
EU – expected utility
3.85
85
Distortions In Deriving Preferences
- Diminishing Sensitivity
People behave as if the subjective value of an amount is
determined, at least in part, by its rank position in the set
of values currently in a person’s head.
So, for example, $10 has a higher subjective value in the
set $2, $5, $8, and $15 because it ranks 2nd, but has a
lower subjective value in the set $2, $15, $19, and $25
because it ranks 4th.
This suggestion — that subjective value is rank within a
sample — is consistent with Parducci’s (1965, 1995) rangefrequency model of magnitudes.
3.86
86
Distortions In Deriving Preferences
- Diminishing Sensitivity
Booij and van de Kuilen (2009) present an experiment that
completely measures the utility - and loss aversion
component of risk attitudes, using a representative sample
of N = 1935 respondents from the general public, in a
parameter-free way. The study is valid under (cumulative)
prospect theory, does not depend on prior assumptions
about the underlying functional form of utility, is externally
valid, and does not rule out heterogeneity of individual
preferences. The results confirm the concave–convex
pattern of utility as predicted by prospect theory, suggest
that utility curvature is less pronounced than suggested by
classical utility measurements, and show that women are
significantly more loss averse than men.
3.87
87
Distortions In Deriving Preferences
- Diminishing Sensitivity
See the next slide for clearer plots
Where low/high stimuli reflect the financial levels of
3.88
88
the decision the participants had to make.
Distortions In Deriving Preferences
- Diminishing Sensitivity
1
0.8
0.6
0.4
0.2
U
low-stimuli
0
-1000
-0.2
0
1000
2000
3000
4000
high-stimuli
-0.4
-0.6
-0.8
-1
x
3.89
89
What Might Utility Look Like?
The Rieskamp
(2008) experiement,
180 trials on 30
individuals.
My interpretation
3.90
90
What Might Utility Look Like?
-100
y1
1.0
1.0
0.5
0
y2
0.5
0.0
1.0
0.5
1.0
0.5
1.0
0.5
0.0
-100
0
100
1.0
0.0
y 11
1.0
1.0
0.5
0.0
0.0
y 12
0.5
0.0
y 15
1.0
0.5
y8
0.5
0.0
y 14
100
0.0
y7
1.0
0.5
0.0
y 13
1.0
0
y4
0.5
0.0
y 10
0.5
0.0
1.0
0.5
0.0
y9
1.0
1.0
0.0
y6
0.5
0.0
-100
y3
0.5
0.0
y5
1.0
100
-100
0
100
These are my numerical solutions to multiple choice
experiments and may not always be a feasible solution
(see case 22 on the module web site).
3.91
91
What Might Utility Look Like?
-100
y1
1.0
1.0
0.5
0
y2
0.5
0.0
1.0
1.0
0.5
1.0
0.5
1.0
0.5
0.0
-100
0
100
1.0
1.0
0.0
y 11
1.0
1.0
0.5
0.0
0.0
y 12
0.5
0.0
y 15
1.0
0.5
y8
0.5
0.0
y 14
100
0.0
y7
0.5
0.0
y 13
1.0
0
y4
0.5
0.0
y 10
0.5
0.0
1.0
0.5
0.0
y9
1.0
1.0
0.0
y6
0.5
0.0
-100
y3
0.5
0.0
y5
100
-100
0
100
For more details see Rieskamp, J. 2008 “The probabilistic nature of
preferential choice”, Journal Of Experimental Psychology. Learning,
Memory, And Cognition, 34, 1446-65 also refer to Abdellaoui, M. 2000
“Parameter-free elicitation of utility and probability weighting
functions” Management Science 46(11) 1497-1512 and later works by
3.92
92
this author.
What Might Utility Look Like?
However, beware “Thus, the data imply that despite
their historical importance and incorporation in many
psychological and economic decision theories, the
most widely assumed models of utility are incorrect”
(Kirby 2011). For a nice, relatively simple,
mathematical introduction see Špirková(2013), who
aims to calibrate utility using a short questionnaire.
For a theoretical account of the origin of the shapes
of utility functions, see Stewart et al. (2014).
3.93
93
What Might Utility Look Like?
Malul et al. (2013) describe three different
experiments that explore participants’ risk attitude.
When they analysed the average results, they found
that participants behave as the S-shape value
function predicts. However, breaking the data down
on the individual level reveals that the S-shape is
valid just for about one-third of the cases.
3.94
94
What Might Utility Look Like?
In the first experiment, they used lotteries with
different stakes and found that in the high stake
only 31% of the participants behave as the S-shape
value function predicts. The percentage decreases to
16% when the stakes were lowered. In the second
experiment, they used a prepayment mechanism to
create a more realistic experimental environment. In
this case, 37% of the participants behaved
consistently with the S-shape value function. In the
third experiment, they used allocation tasks. The
results revealed that most subjects could not be
classified into one of the classical risk attitude
groups. More than one value function is needed to
3.95
95
characterize individuals’ attitudes toward risk.
What Might Utility Look Like?
There might also be wider applications, Hsu and
Vlaev (2014) measured utility curves for the
hypothetical monetary costs as a function of time
engaged in three everyday physical activities:
walking, standing, and sitting. They found that
activities requiring more physical exertion resulted
in steeper discount curves, i.e., perceived cost as a
function of time. They also examined the effects of
gain versus loss framing (whether the activity
brought additional rewards or prevented losses) as
well as the effects of the individual factors of
gender, income, and BMI. The results also
demonstrate a general method for examining the
96
costs of effort associated with everyday activities.3.96
What Might Utility Look Like?
It has been said “Essentially, all models are wrong,
but some are useful” G.E.P. Box (Box, G.E.P., and
Draper, N.R., 1987 page 424).
3.97
97
Cumulative Prospect Theory
For many years this has been a useful descriptive
model. In recent years a more complex alternative
has been introduced (Tversky and Kahneman, 1992)
now called the Cumulative Prospect Theory. Much
recent work has investigated this problem; in
particular obtaining fits to experimental data.
See case 22 on the module web site.
3.98
98
Distortions In Deriving
Preferences - Altruism
Simple altruism does parsimoniously capture
important phenomena, and is psychologically valid in
many contexts.
But there is a mass of psychological evidence - and,
more recently, experimental economic evidence –
that indicates it is often an importantly wrong model
of social preferences.
3.99
99
Distortions In Deriving
Preferences - Altruism
To get a sense for how social preferences differ
from simple altruism, I turn to a (far from complete)
account of what can be called “behavioural
distributive justice”: How do individuals choose to
divide resources among themselves and others?
There are two aspects to this question: First, what
do people, when disinterested, feel are proper rules
for allocation? Second, to what degree do people
sacrifice self-interest for the sake of these
principles?
3.100
100
Distortions In Deriving
Preferences - Altruism
To address the question of disinterested
assessments, suppose two people together find $10
on the ground.
How would the average person, in their role as a
third party, decide to split the money between the
two?
One answer, following from the simple-altruism
perspective, is that the person who is poorer, or who
can otherwise benefit most from the money, should
get it.
3.101
101
Distortions In Deriving
Preferences - Altruism
There is no doubt that people often consider
comparative needs in allocation decisions.
There are similar norms about how to allocate goods
whose usefulness is different for the two parties:
We often find it appealing to allocate goods to
maximize the benefit of each good.
3.102
102
Distortions In Deriving Preferences
- Reciprocity and Attribution
The previous subsection (see extend example 3)
considered evidence about social preferences
defined over the allocations of goods.
Psychological evidence indicates, however, that social
preferences are not merely a function of
consumption levels, or even changes in consumption
levels: People do not seek uniformly to help other
people, nor do they seek uniformly to share
equitably.
3.103
103
Distortions In Deriving Preferences
- Reciprocity and Attribution
Rather, people’s concerns regarding the experiences of
other people depend on the behaviour, motivations, and
intentions of these other people.
The same people who are altruistic towards deserving
people are often indifferent to the plight of
undeserving people, and often motivated to hurt those
whom they believe to have behaved egregiously.
If somebody is being nice to you or others, you are
inclined to be nice to him. If somebody is being mean to
you or others, you are inclined to be mean to them.
3.104
104
Distortions In Deriving Preferences
- Reciprocity and Attribution
This “reciprocal” nature of preferences manifests
itself in the distinction between simple altruism, as
outlined earlier, and reciprocal altruism.
Consider the question of why people conserve water
during a drought. Clearly they perceive that
conservation contributes to the general good, which
at a small cost is something they eagerly do. How
might we model such preferences?
3.105
105
Distortions In Deriving Preferences
- Reciprocity and Attribution
First note that probably there are diminishing social
benefits of conservation because the marginal social
value of water is greater the less water there is.
If other people conserve, it is less urgent for you to
do so; if other people don’t conserve, it is more urgent
for you to do so.
If you were a simple altruist, therefore, learning that
others were not conserving should cause you to
intensify your conservation efforts — if nobody else
conserves, it becomes all the more urgent that you do.
3.106
106
Distortions In Deriving Preferences
- Reciprocity and Attribution
This prediction does not accord with actual
sentiments toward cooperation. People are more
inclined to conserve energy or water if they think
other people are doing their share, but not if they
think that others are not doing their part. People
reciprocate public spiritedness in others rather than
counteract it.
3.107
107
More On Altruism
Almost 10 million NHS admissions in England related
to alcohol last year - Mirror - 15 Oct 2014
Heavy boozers are putting the NHS under
“intolerable strain” and risk sparking a health crisis
which will cost the country billions.
Alcohol Concern said 9.9million NHS admissions in
England – including hospital patients and clinic and
A&E visits – were related to alcohol last year.
3.108
108
More On Altruism
Some 9.6 million people are now drinking in excess of
Government guidelines - including 2.4 million who are
classed as “high risk”, according to the charity.
High risk drinking is defined as people who drink
more than six to eight units of alcohol a day, with one
unit equating to less than a small glass of wine or a
half pint of beer.
The charity’s chief executive, Jackie Ballard, said:
“The NHS is now facing an intolerable strain from
alcohol-related illnesses."
3.109
109
Distortions In Deriving Preferences
– Prisoner’s Dilemma
A scenario where cooperation and trust wins and blind
pursuit of self-interest loses, is illustrated by the problem
faced by two accomplices locked in separate cells. Each is
offered three choices by the police:
(1)
if both confess to the charges, both will be jailed
for five years
(2)
if only one confesses, he will be freed but the nonconfessor will be jailed for twenty years
(3)
if neither confesses, both will be tried for a minor
offence and will be jailed for one year
3.110
110
Distortions In Deriving Preferences
– Prisoner’s Dilemma
vvvvvvvvvvvvvvvv
3.111
111
Distortions In Deriving Preferences
– Prisoner’s Dilemma
If both know that the other will not be selfish and will
take the collective interest into consideration, neither will
confess and serve one year in jail.
Otherwise, where one cannot depend on the other, both
have no choice but to confess and serve five years. It is an
example of non-zero sum game.
3.112
112
Distortions In Deriving Preferences
- Reciprocity and Attribution
The results of a competition to devise a computer
programme capable of optimally solving the prisoner’s
dilemma problem are reported (Axelrod and Hamilton
1981, Axelrod 1984). The winning programme always
initially uses a co-operative strategy until the opposing
player defects and then retaliates until the opponent cooperates again.
Their “model is based on the more realistic assumption
that the number of interactions is not fixed in advance.
Instead, there is some probability, that after the current
interaction the same two individuals will meet again.
Factors that affect the magnitude of this probability of
meeting again include the average lifespan, relative
3.113
113
mobility, and health of the individuals.”
Distortions In Deriving Preferences
- Reciprocity and Attribution
This ‘tit-for-tat’ strategy turns out to be optimal almost
whatever the opponent does (Hertwig and Todd,
Gigerenzer 2007 and Gigerenzer 2008). If my opponent is
continually uncooperative he gets one period of grace
before my retaliation kicks in. If my opponent always cooperates so will I. Interestingly the strategy that
performed worst is the most seemingly ‘‘sophisticated’’
invoking learning and probability distributions to
constantly update behaviour.
3.114
114
Distortions In Deriving Preferences
- Reciprocity and Attribution
The most direct evidence of reciprocal altruism in the
prisoner’s dilemma with which I am familiar comes from
Shafir and Tversky (1992).
When subjects were told that their anonymous partner
in a Prisoners’ Dilemma had cooperated, 16% also
cooperated; when subjects were told that their partner
did not cooperate, only 3% cooperated.
This idea that people are motivated to retaliate when
they feel they have been mistreated is a fairly obvious
intuition, and well understood by psychologists.
3.115
115
Distortions In Deriving Preferences
- Reciprocity and Attribution
More recently, it has been widely explored by
experimentalists who have investigated many variants of
the “ultimatum game” (Mousazadeha and Izadkhahb 2015).
The ultimatum game consists of two people splitting some
fixed amount of money according to the following rules:
A Proposer offers some division of (say) $10 to a Decider.
If the Decider accepts, they split the money according to
the proposal.
If the Decider rejects, they both get nothing.
3.116
116
Distortions In Deriving Preferences
- Reciprocity and Attribution
The result of rational self-interest is clear:
Proposers will never offer more than a penny, and
the Decider will accept any offer of at least a penny.
Experiments clearly refute such behaviour: Even in
one-shot settings, Deciders are willing to punish
unfair offers by rejecting them, and Proposers tend
to make fair offers.
3.117
117
Distortions In Deriving Preferences
- Reciprocity and Attribution
The decision by Proposers to make fair offers can
come from at least two motivations:
Proposers themselves may have a preference for
being fair,
or self-interested Proposers might correctly predict
that Deciders will retaliate against unfair offers by
rejecting them.
3.118
118
Distortions In Deriving Preferences
- Reciprocity and Attribution
Criminals show cooperation and prosocial behaviour in
economic games. It's easy to demonise people who have
broken the law. However, recent studies using economic
games that test fairness and cooperation show that this is
short-sighted. Researchers observed prisoners'
performance on a famous game known as the "prisoner's
dilemma" (Khadjavi and Lange 2013) - the convicted
criminals actually displayed more cooperation during the
game than undergraduate students. Similarly, another study
(Birkeland et al. 2014) found that people with a criminal
record displayed just as much "prosocial motivation" (i.e.
they distributed money fairly) in the "dictator game" (see
3.119
119
below) as those without such a record.
Distortions In Deriving Preferences
- Reciprocity and Attribution
For those who wish to put the problem into practice
the Betrayer's Banquet is an evening of theatrical
dining crossed with the prisoner's dilemma. Mind you
at around £100 a ticket it is not a cheap experience.
How the prisoner's dilemma changes diners'
etiquette - physics-math - 26 September 2013 - New
Scientist
Betrayers' Banquet
3.120
120
Distortions In Deriving Preferences
- Reciprocity and Attribution
In a similar vein consider the “dictator game” as
reviewed by Krupka and Weber (2013) and the
“ultimatum game” reviewed by Mousazadeha and
Izadkhahb (2015).
For an easy reading introduction see Pauli (2009).
3.121
121
Distortions In Deriving
Preferences - Prospect Theory
The distortions in this section are often modelled
using the prospect theory proposed in Kahneman and
Tversky (1979), which suitably modifies the
expected utility formulation.
The major modifications are three.
3.122
122
Distortions In Deriving
Preferences - Prospect Theory
First, the utility function is defined over changes in
wealth rather than wealth levels.
Second, the slope of the utility function over
changes in wealth is greater for losses than for
gains.
Third, the agent generates decision weights wi from
the probability distribution and maximizes Σwiu(xi).
Note that the decision weights are not necessarily
interpretable as probability weights.
3.123
123
Distortions In Deriving
Preferences - Prospect Theory
This basic model is often enriched by the assumption
that the utility function is concave over gains and
convex over losses: people are risk averse when
dealing with gains and risk prone when dealing with
loses.
3.124
124
Distortions In Deriving
Preferences - Prospect Theory
In his review of 30 years of research in Prospect Theory,
Barberis (2013) notes that support for Prospect Theory had
come mainly from the laboratory. In a paper (Abdel-khalik
2014), writes about recurring phenomenon in real life that are
consistent with Prospect Theory predictions in the decisionmaking loss domain. The 58 cases noted in his paper are
associated with specific risk seekers that had cost more than
$126 billion (an average of $2.3 billion per case). Synopses are
presented for 14 cases. It is striking that these cases are
costly, all participants are young men, and almost all had
followed the gambler’s martingale strategy — i.e., double down.
While these cases are informative about risk-seeking
behaviour, they are not sufficiently systematic to be subjected
to stylized archival research methods.
3.125
125
Prospect Theory - Decision Weights
This discussion closely follows that of Kahneman
(2011). Many years after they published prospect
theory, Tversky and Kahneman (1992), carried out a
study in which they measured the decision weights
that explained people’s preferences for gambles with
modest monetary stakes. The estimates for gains are
shown in the table and graph.
probability
weight
probability
weight
0
0
1
5.5
80
60.1
2
8.1
90
71.2
5
13.2
95
79.3
10
18.6
98
87.1
20
26.1
99
91.2
50
42.1
100
100
3.126
126
Prospect Theory - Decision Weights
The estimates for gains are shown in the previous
table and graph below.
100
90
80
weight
70
60
50
40
30
20
10
0
0
10
20
30
40
50
60
proba bility
70
80
90
100
3.127
127
Prospect Theory - Decision Weights
The particular probability-weighting function originally proposed
by Tversky and Kakneman (1992) is
p
( p ) 
0   1
1


p  (1  p ) 
where p is the cumulative probability of the distribution of gains
and losses. In this case the value fitted to the experimental data
is   0.61. The weighting function has an inverted S-shape and is
sub-proportional. That is if for p > q and 0    1 then
( q ) ( p )

( q )
( p )


The property allows for the Allais paradox (lecture 7) and can
explain the preference for lottery tickets (Ingersoll 2008).
3.128
128
Prospect Theory - Decision Weights
For a review of several different probability weighting
functions, plus a cautionary discussion see Ingersoll
(2008). Köbberling and Wakker (2003) apply the
tradeoff technique to three popular theories of
individual decision under uncertainty and risk, i.e.,
expected utility, Choquet expected utility (Etner et al.
2012), and prospect theory.
3.129
129
Prospect Theory - Decision Weights
probability
weight
probability
weight
0
0
1
5.5
80
60.1
2
8.1
90
71.2
5
13.2
95
79.3
10
18.6
98
87.1
20
26.1
99
91.2
50
42.1
100
100
You can see that the decision weights are identical to the
corresponding probabilities at the extremes: both equal to
0 when the outcome is impossible, and both equal to 100
when the outcome is a sure thing. However, decision
weights depart sharply from probabilities near these
points. At the low end, we find the possibility effect:
unlikely events are considerably over weighted.
3.130
130
Prospect Theory - Decision Weights
probability
weight
probability
weight
0
0
1
5.5
80
60.1
2
8.1
90
71.2
5
13.2
95
79.3
10
18.6
98
87.1
20
26.1
99
91.2
50
42.1
100
100
For example, the decision weight that corresponds to a 2%
chance is 8.1. If people conformed to the axioms of
rational choice, the decision weight would be 2 — so the
rare event is over weighted by a factor of 4. The certainty
effect at the other end of the probability scale is even
more striking. A 2% risk of not winning the prize reduces
the utility of the gamble by 13%, from 100 to 87.1.
3.131
131
Prospect Theory - Decision Weights
To appreciate the asymmetry between the possibility
effect and the certainty effect, imagine first that you
have a 1% chance to win $1 million. You will know the
outcome tomorrow. Now, imagine that you are almost
certain to win $1 million, but there is a 1% chance that you
will not. Again, you will learn the outcome tomorrow. The
anxiety of the second situation appears to be more salient
than the hope in the first. The certainty effect is also
more striking than the possibility effect if the outcome is
a surgical disaster rather than a financial gain. Compare
the intensity with which you focus on the faint sliver of
hope in an operation that is almost certain to be fatal,
compared to the fear of a 1% risk.
3.132
132
Prospect Theory - Decision Weights
probability
weight
probability
weight
0
0
1
5.5
80
60.1
2
8.1
90
71.2
5
13.2
95
79.3
10
18.6
98
87.1
20
26.1
99
91.2
50
42.1
100
100
The combination of the certainty effect and possibility
effects at the two ends of the probability scale is
inevitably accompanied by inadequate sensitivity to
intermediate probabilities. You can see that the range of
probabilities between 5% and 95% is associated with a
much smaller range of decision weights (from 13.2 to
79.3), about two-thirds as much as rationally expected.
3.133
133
Prospect Theory - Decision Weights
Neuroscientists (Hsu et al. 2009) have confirmed these
observations, finding regions of the brain that respond to
changes in the probability of winning a prize. The brain’s
response to variations of probabilities is strikingly similar
to the decision weights estimated from choices.
3.134
134
Prospect Theory - Decision Weights
Probabilities that are extremely low or high (below 1% or
above 99%) are a special case. It is difficult to assign a
unique decision weight to very rare events, because they
are sometimes ignored altogether, effectively assigned a
decision weight of zero. On the other hand, when you do
not ignore the very rare events, you will certainly
overweight them. Most of us spend very little time
worrying about nuclear meltdowns or fantasizing about
large inheritances from unknown relatives.
3.135
135
Prospect Theory - Decision Weights
However, when an unlikely event becomes the focus of
attention, we will assign it much more weight than its
probability deserves. Furthermore, people are almost
completely insensitive to variations of risk among small
probabilities. A cancer risk of 0.001% is not easily
distinguished from a risk of 0.00001%, although the
former would translate to 3,000 cancers for the
population of the United States, and the latter to 30.
3.136
136
Prospect Theory - Decision Weights
When you pay attention to a threat, you worry — and the
decision weights reflect how much you worry. Because of
the possibility effect, the worry is not proportional to the
probability of the threat. Reducing or mitigating the risk is
not adequate; to eliminate the worry the probability must
be brought down to zero.
3.137
137
Prospect Theory - Decision Weights
The question below is adapted from a study of the
rationality of consumer valuations of health risks (Viscusi
et al. 1987), which was published by a team of economists
in the 1980s. The survey was addressed to parents of small
children.
Suppose that you currently use an insect spray that costs
you $10 per bottle and it results in 15 inhalation poisonings
and 15 child poisonings for every 10,000 bottles of insect
spray that are used. You learn of a more expensive
insecticide that reduces each of the risks to 5 for every
10,000 bottles. How much would you be willing to pay for
it?
3.138
138
Prospect Theory - Decision Weights
Parents were willing to pay an additional $2.38, on average,
to reduce the risks by two-thirds from 15 per 10,000
bottles to 5. They were willing to pay $8.09, more than
three times as much, to eliminate it completely. Other
questions showed that the parents treated the two risks
(inhalation and child poisoning) as separate worries and
were willing to pay a certainty premium for the complete
elimination of either one. This premium is compatible with
the psychology of worry but not with the rational model.
3.139
139
The Fourfold Pattern
This discussion closely follows that of Kahneman (2011).
When Tversky and Kahneman (1992) began work on
prospect theory, they quickly reached two conclusions:
people attach values to gains and losses rather than to
wealth, and the decision weights that they assign to
outcomes are different from probabilities. Neither idea
was completely new, but in combination they explained a
distinctive pattern of preferences that we called the
fourfold pattern. The name has stuck. The scenarios are
illustrated below.
3.140
140
The Fourfold Pattern
GAINS
LOSSES
HIGH
95% chance to win $10,000
95% chance to lose $10,000
PROBABILITY
Fear of disappointment
Hope to avoid loss
Certainty Effect
RISK AVERSE
RISK SEEKING
Accept unfavourable settlement
Reject favourable settlement
LOW
5% chance to win $10,000
5% chance to lose $10,000
PROBABILITY
Hope of large gain
Fear of large loss
Possibility Effect
RISK SEEKING
RISK AVERSE
Reject favourable settlement Accept unfavourable settlement
The top row in each cell shows an illustrative prospect.
3.141
141
The Fourfold Pattern
GAINS
LOSSES
HIGH
95% chance to win $10,000
95% chance to lose $10,000
PROBABILITY
Fear of disappointment
Hope to avoid loss
Certainty Effect
RISK AVERSE
RISK SEEKING
Accept unfavourable settlement
Reject favourable settlement
LOW
5% chance to win $10,000
5% chance to lose $10,000
PROBABILITY
Hope of large gain
Fear of large loss
Possibility Effect
RISK SEEKING
RISK AVERSE
Reject favourable settlement Accept unfavourable settlement
The second row characterizes the focal emotion that the
prospect evokes.
3.142
142
The Fourfold Pattern
GAINS
LOSSES
HIGH
95% chance to win $10,000
95% chance to lose $10,000
PROBABILITY
Fear of disappointment
Hope to avoid loss
Certainty Effect
RISK AVERSE
RISK SEEKING
Accept unfavourable settlement
Reject favourable settlement
LOW
5% chance to win $10,000
5% chance to lose $10,000
PROBABILITY
Hope of large gain
Fear of large loss
Possibility Effect
RISK SEEKING
RISK AVERSE
Reject favourable settlement Accept unfavourable settlement
The third row indicates how most people behave when offered a
choice between a gamble and a sure gain (or loss) that corresponds to
its expected value (for example, between “95% chance to win $10,000”
and “$9,500 with certainty”). Choices are said to be risk averse if the
sure thing is preferred, risk seeking if the gamble is preferred.
3.143
143
The Fourfold Pattern
GAINS
LOSSES
HIGH
95% chance to win $10,000
95% chance to lose $10,000
PROBABILITY
Fear of disappointment
Hope to avoid loss
Certainty Effect
RISK AVERSE
RISK SEEKING
Accept unfavourable settlement
Reject favourable settlement
LOW
5% chance to win $10,000
5% chance to lose $10,000
PROBABILITY
Hope of large gain
Fear of large loss
Possibility Effect
RISK SEEKING
RISK AVERSE
Reject favourable settlement Accept unfavourable settlement
The fourth row describes the expected attitudes of a defendant
and a plaintiff as they discuss a settlement of a civil suit.
3.144
144
The Fourfold Pattern
The fourfold pattern of preferences is considered one of
the core achievements of prospect theory. Three of the
four cells are familiar; the fourth (top right) was new and
unexpected.
3.145
145
The Fourfold Pattern
The top left is the one that Bernoulli (1700–1782)
discussed: people are averse to risk when they consider
prospects with a substantial chance to achieve a large gain.
They are willing to accept less than the expected value of
a gamble to lock in a sure gain.
3.146
146
The Fourfold Pattern
The possibility effect in the bottom left cell explains why
lotteries are popular. When the top prize is very large,
ticket buyers appear indifferent to the fact that their
chance of winning is minuscule. A lottery ticket is the
ultimate example of the possibility effect. Without a
ticket you cannot win, with a ticket you have a chance, and
whether the chance is tiny or merely small matters little.
Of course, what people acquire with a ticket is more than a
chance to win; it is the right to dream pleasantly of
winning.
3.147
147
The Fourfold Pattern
The bottom right cell is where insurance is bought. People
are willing to pay much more for insurance than expected
value — which is how insurance companies cover their costs
and make their profits. Here again, people buy more than
protection against an unlikely disaster; they eliminate a
worry and purchase peace of mind.
3.148
148
The Fourfold Pattern
The results for the top right cell initially surprised
Kahneman and Tversky.
3.149
149
The Fourfold Pattern
The results for the top right cell initially surprised
Kahneman and Tversky. They were accustomed to think in
terms of risk aversion except for the bottom left cell,
where lotteries are preferred. When they looked at the
choices for bad options, they quickly realized that
individuals were just as risk seeking in the domain of
losses, as they were risk averse in the domain of gains.
They were not the first to observe risk seeking with
negative prospects — at least two authors had reported
that fact (Markowitz 1952 and Williams 1966 and further
discussed in Kahneman and Tversky 1979), but they had
not made much of it. However, Tversky and Kahneman
(1992) were fortunate to have a framework that made the
finding of risk seeking easy to interpret, and that was a
milestone in their thinking. Indeed, they identified two
3.150
150
reasons for this effect.
The Fourfold Pattern
First, there is diminishing sensitivity. The sure loss is very aversive
because the reaction to a loss of $900 is more than 90% as intense as
the reaction to a loss of $1,000.
The second factor may be even more powerful: the decision weight
that corresponds to a probability of 90% is only about 71 (refer back
to the table of weights), much lower than the probability. The result is
that when you consider a choice between a sure loss and a gamble with
a high probability of a larger loss, diminishing sensitivity makes the
sure loss more aversive, and the certainty effect reduces the
aversiveness of the gamble.
The same two factors enhance the attractiveness of the sure thing
and reduce the attractiveness of the gamble when the outcomes are
positive.
3.151
151
The Fourfold Pattern
The shape of the value function and the decision weights both
contribute to the pattern observed in the top row of the table. In the
bottom row, however, the two factors operate in opposite directions:
diminishing sensitivity continues to favour risk aversion for gains and
risk seeking for losses, but the over weighting of low probabilities
overcomes this effect and produces the observed pattern of gambling
for gains and caution for losses.
3.152
152
The Fourfold Pattern
Many unfortunate human situations unfold in the top right
cell. This is where people who face very bad options take
desperate gambles, accepting a high probability of making
things worse in exchange for a small hope of avoiding a
large loss. Risk taking of this kind often turns manageable
failures into disasters. The thought of accepting the large
sure loss is too painful, and the hope of complete relief too
enticing, to make the sensible decision that it is time to
cut one’s losses. This is where businesses that are losing
ground to a superior technology waste their remaining
assets in futile attempts to catch up. Because defeat is so
difficult to accept, the losing side in wars often fights
long past the point at which the victory of the other side
is certain, and only a matter of time.
3.153
153
Distortions In Deriving Preferences Shape And Attractiveness
The following table list eight gambles in Euros as
ranked by financial analysts according to their
attractiveness to investors. They have the same
expected value and the same number of possible
outcomes (Kahneman and Riepe 1998 ).
Which would you prefer?
3.154
154
Distortions In Deriving Preferences Shape And Attractiveness
Gamble
Payoff 1
Probability (%)
Payoff 2
Probability (%)
A
5,000
95
105,000
5
B
5,000
50
15,000
50
C
1,000
10
11,000
90
D
1,000
90
91,000
10
E
2,000
50
18,000
50
F
0
50
20,000
50
G
-2,000
90
118,000
10
H
-5,000
50
25,000
50
3.155
155
Distortions In Deriving Preferences Shape And Attractiveness
Gamble
Payoff 1
Probability (%)
Payoff 2
Probability (%)
A
5,000
95
105,000
5
B
5,000
50
15,000
50
C
1,000
10
11,000
90
F
0
50
20,000
50
G
-2,000
90
118,000
10
H
-5,000
50
25,000
50
The ideal gamble combines a high probability of a
1,000
90
91,000
10
D
moderate gain and a small probability of a very large
2,000
50
18,000
50
gain.E
3.156
156
Distortions In Deriving Preferences Shape And Attractiveness
Gamble
Payoff 1
Probability (%)
Payoff 2
Probability (%)
A
5,000
95
105,000
5
B
5,000
50
15,000
50
C
1,000
10
11,000
90
G
-2,000
90
118,000
10
H
-5,000
50
25,000
50
ItsDprefferability
may be
by a combination
1,000
90 explained
91,000
10
of the over weighting of the small probability of a
2,000
50
18,000
50
E
large gain and of the different risk attitudes over
0
50
20,000
50
F and losses.
gains
3.157
157
Distortions In Deriving Preferences Shape And Attractiveness
Gamble
Payoff 1
Probability (%)
Payoff 2
Probability (%)
A
5,000
95
105,000
5
B
5,000
50
15,000
50
C
1,000
10
11,000
90
D
1,000
90
91,000
10
E
2,000
50
18,000
50
F
0
50
20,000
50
G
-2,000
90
118,000
10
H
-5,000
50
25,000
50
3.158
158
How Do You Choose An Analyst?
The idea that financial analysts play an important role in
financial markets is rather consensual (Cowles, 1933;
O’Brien, 1990). Yet there is some debate on whether
following the advice of analysts brings value to investors
after transaction costs (Womack, 1996; Mikhail,
Walther, and Willis, 2004; Li, 2005). Related to this is
the difficulty in identifying the analysts with superior
stock picking skills. In a paper Aiguzhinov et al. (2015)
show that the rankings of financial analysts are useful to
investors because strategies based upon these rankings
yield positive abnormal returns.
3.159
159
Next Week
Framing Effects
3.160
160