ORGANIZATIONAL BEHAVIOR AND HUMAN DECISION PROCESSES 35, 124-140(1985)
The Psychology of Sunk Cost
HAL R. ARKES AND CATHERINE BLUMER
Ohio University
The sunk cost effect is manifested in a greater tendency to continue an
endeavor once an investment in money, effort, or time has been made. Evidence
that the psychological justification for this behavior is predicated on
the desire not to appear wasteful is presented. In a field study, customers who
had initially paid more for a season subscription to a theater series attended
more plays during the next 6 months, presumably because of their higher sunk
cost in the season tickets. Several questionnaire studies corroborated and
extended this finding. It is found that those who had incurred a sunk cost
inflated their estimate of how likely a project was to succeed compared to the
estimates of the same project by those who had not incurred a sunk cost. The
basic sunk cost finding that people will throw good money after bad appears
to be well described by prospect theory (D. Kahneman & A. Tversky, 1979,
Econometrica, 47, 263-291). Only moderate support for the contention that
personal involvement increases the sunk cost effect is presented. The sunk
cost effect was not lessened by having taken prior courses in economics.
Finally, the sunk cost effect cannot be fully subsumed under any of several
social psychological theories. 0 1985 Academic PRSS, IIIC.
To terminate a project in which $1.1 billion has been invested represents an unconscionable
mishandling of taxpayers’ dollars.
Senator Denton, November 4, 1981
Completing Tennessee-Tombigbee [Waterway Project] is not a waste of taxpayer
dollars. Terminating the project at this late stage of development would, however,
represent a serious waste of funds already invested.
Senator Sasser, November 4, 1981
The purpose of the present paper is to attempt to explain an irrational
economic behavior, which will be termed the sunk cost effect. This effect
is manifested in a greater tendency to continue an endeavor once an
investment in money, effort, or time has been made. The prior investment,
which is motivating the present decision to continue, does so despite
the fact that it objectively should not influence the decision. We will
We thank Robert Winter and Mindy Sherbs for their cooperation and assistance in the
theater experiment. That study was supported by a grant from the Ohio University Research
Committee. We also thank several economists who provided us with valuable comments
and criticisms: Arnold Katz, David Klingaman, Jack Ochs, and Richard Vedder. Requests
for reprints should be sent to Hal R. Arkes, Department of Psychology, Ohio University,
Athens, OH 45701.
124
0749-5978185 $3.00
Copyright Q 1985 by Academic Press, Inc.
All rights of reproduction in any form reserved.
SUNK COST 125
provide evidence that the psychological justification for this maladaptive
behavior is predicated on the desire not to appear wasteful.
As an example of the sunk cost effect, consider the following example. ’
A man wins a contest sponsored by a local radio station. He is given a
free ticket to a football game. Since he does not want to go alone, he
persuades a friend to buy a ticket and go with him. As they prepare to
go to the game, a terrible blizzard begins. The contest winner peers out
his window over the arctic scene and announces that he is not going,
because the pain of enduring the snowstorm would be greater than the
enjoyment he would derive from watching the game. However, his friend
protests, “I don’t want to waste the twelve dollars I paid for the ticket!
I want to go!” The friend who purchased the ticket is not behaving rationally
according to traditional economic theory. Only incremental costs
should influence decisions, not sunk costs. If the agony of sitting in a
blinding snowstorm for 3 h is greater than the enjoyment one would derive
from trying to see the game, then one should not go. The $12 has been
paid whether one goes or not. It is a sunk cost. It should in no way
influence the decision to go. But who among us is so rational?
Examples of the sunk cost effect exist in great quantity and for great
quantities. During late 1981 the funding for the immensely expensive
Tennessee-Tombigbee Waterway Project was scheduled for Congressional
review. As the above quotes indicate, proponents of the project
insisted that to stop the project after a great deal had already been spent
would represent a waste of taxpayers’ money. In other words, the sunk
cost provided a strong impetus to continue the project.
Those who are aware of the fact that sunk costs are difficult to ignore
can turn this realization to their advantage. When discussing why he
thought the nuclear energy program would prevail, one nuclear industry
executive explained:
When it comes down to it, no one with any sense would abort a $2.5 billion
construction project. And, by extension, no administration would abort a $200
billion national investment in nuclear energy. So the trick for the industry is to get
more new plants under construction without the (anti-nuclear) movement knowing
about it. By the time they get around to demonstrating and challenging the license,
we’ll have a million tons of steel and concrete in the ground, and no one in their
right mind will stop us (Dowie, 1981, p. 23).
The executive’s final assertion may be correct if what he means by
“right mind” is a typical line of reasoning. However, such reasoning is
’ This example is adapted from Thaler (1980).
126 ARKES AND BLUMER
irrational, no matter how compelling it may seem. To repeat: sunk costs
are irrelevant to current decisions.*
These economic examples should not obscure the fact that there are
numerous nonmonetary sunk costs. Should I continue this unhappy relationship?
I have already put so much into it. Should I continue with
this terrible job? I spent a year in training to get this position. We suspect
that many bad movies are seen to their completion simply because once
the viewer realizes how poor the movie is, several minutes and dollars
have already been invested. This sunk cost promotes lingering until the
bitter end. During the Viet Nam War some people counseled against
ending the hostilities before total victory had been achieved because to
do so would have meant the waste of those lives already lost. Teger (1980)
summarized this phenomenon by suggesting that we often feel we have
too much invested to quit.
Our analysis of the sunk cost effect will be presented in three stages.
First, demonstrations of the effect will be presented. Second, some possible
explanations of the effect will be offered. Third, the relation between
the sunk cost effect and several areas of social psychological research
will be examined.
DEMONSTRATIONS OF THE SUNK COST EFFECT
AI1 of our questionnaire studies were done with Ohio and Oregon college
students as subjects. No subject responded to more than one question.
The number of subjects giving each answer is indicated after every
question.
Experiment 1
Assume that you have spent $100 on a ticket for a weekend ski trip to Michigan.
Several weeks later you buy a $50 ticket for a weekend ski trip to Wisconsin. You
think you will enjoy the Wisconsin ski trip more than the Michigan ski trip. As
you are putting your just-purchased Wisconsin ski trip ticket in your wallet, you
notice that the Michigan ski trip and the Wisconsin ski trip are for the same
weekend! It’s too late to sell either ticket, and you cannot return either one. You
must use one ticket and not the other. Which ski trip will you go on?
$100 ski trip to Michigan 33
$50 ski trip to Wisconsin 28
2We are aware that the senators who supported the Tennessee-Tombigbee Waterway
Project and those who supported or opposed nuclear power may have had motives other
than sunk cost considerations. It is instructive that Senators Denton and Sasser used sunk
cost arguments in defending the project’s continuation. Apparently they felt that sunk cost
was a more compelling rationale than others that might have been offered.
SUNK COST 127
An axiom of traditional economic theory is that decisions should be
based on the costs and benefits that are expected to arise from the choice
of each option. Based on this axiom we would expect everyone to choose
the trip thought to be more enjoyable-the trip to Wisconsin. However,
only 46% of the subjects chose the Wisconsin trip. The 99% confidence
interval around this datum is 30 to 62%. We therefore conclude that the
prediction of traditional economic theory that 100% of the subjects would
choose the Wisconsin trip is discontirmed. Obviously the larger sunk cost
of the Michigan trip is influencing many subjects’ choice.
Experiment 2
Experiment 1 was a questionnaire study. Actual money was obviously
not involved. While a number of experiments have shown that the results
of questionnaire studies replicated when real monetary stakes were introduced
(e.g., Grether & Plott, 1979; Lichtenstein, 1973), we felt it would
be desirable to demonstrate the sunk cost effect in a more realistic setting.
We decided to provide discounts to some subscribers to a theater series.
We predicted that those who paid less for the privilege to see as many
plays as they liked would choose to see fewer plays than those who had
paid more. Those who had paid more would have a greater sunk cost.
Method. The first 60 people who approached the ticket window to
purchase season tickets to the Ohio University Theater’s 1982-1983
season were included in the experiment. After the person announced his
or her intention to buy a season ticket, the ticket seller sold the purchaser
one of three types of tickets, which had been randomly ordered beforehand.
One type was the normal price ticket ($15); the second was a ticket
selling at a $2 discount; the third was selling at a $7 discount. The seller
explained to the latter two groups that the discount was being given as
part of a promotion by the theater department.
As we sold the three types of tickets, we immediately decided not to
use the forthcoming data from three $7 discount subjects, one $2 discount
subject, and two no-discount subjects. This was because these six people
were buying tickets as couples. If the two members of a couple had tickets
with different discounts, their joint decision whether or not to attend a
play would render their data nonindependent. Our final sample thus had
eighteen no-discount, nineteen $2 discount, and seventeen $7 discount
subjects. Since the ticket stubs were color coded, we were able to collect
the stubs after each performance and determine how many persons in
each group had attended each play. Each season ticket package contained
one ticket labeled with the name of one play plus two unlabeled extra
tickets which could be used to bring a guest to any play.
Results. We divided the theater season into halves-the first five plays
and the last five plays-because we felt that the experimental manipu-
128 ARKES AND BLUMER
lation might be of different strength during the two halves of the season.
We performed a 3 (discount: none, $2, $7) x 2 (half of season) analysis
of variance on the number of tickets used by each subject. The latter
variable was a within-subjects factor. It was also the only significant
source of variance, F(1,51) = 32.32, MS, = 1.81, p < .OOl. More tickets
were used by each subject on the first five plays (3.57) than on the last
five plays (2.09). We performed a priori tests on the number of tickets
used by each of the three groups during the first half of the theater season.
The no-discount group used significantly more tickets (4.11) than both
the $2 discount group (3.32) and the $7 discount group (3.29), t = 1.79,
1.83, respectively, p’s < .05, one tailed. The groups did not use significantly
different numbers of tickets during the last half of the theater
season (2.28, 1.S4, 2.18, for the no-discount, $2 discount, and $7 discount
groups, respectively).
Conclusion. Those who had purchased theater tickets at the normal
price used more theater tickets during the first half of the season than
those who purchased tickets at either of the two discounts. According to
rational economic theory, after all subjects had their ticket booklet in
hand, they should have been equally likely to attend the plays. Since the
discounts were assigned randomly, the groups should not have differed
on the costs and benefits they could have anticipated by attending each
play. The groups did differ, however, because they had different sunk
costs.3
We consider this demonstration of the sunk cost effect to be particularly
noteworthy, because the effect lasted for 6 months following the
purchase of the tickets. The effect was not manifested, however, during
the second half of the theater series (6 to 9 months following the pur-
chase).
Experiments 1 and 2 are relatively pure examples in that other explanations
of the results are not readily available. Many of the following
studies are less pure. They involve much more complex economic decisions
than are required in the first two experiments. As a result of using
more complex stories, we are creating a stimulus situation in which some
explanations of the data other than the sunk cost effect may exist. It is
virtually impossible to rule out every alternate explanation in every such
story. However, the consistent pattern of results found in all of the stories
plus the demonstration of the sunk cost effect in the purer stories lead
us to feel confident in our explanation of the data.
3In this analysis we are ignoring “income effects.” While it is true that the discount
groups had more disposable income with which to seek nontheater entertainment during
the year, we contend that this extra income was not the reason they attended fewer plays.
The $2 discount group, whose members manifested the sunk cost effect, had only an extra
6e of disposable income per week during the course of the experiment.
SUNK COST 129
The next three experiments differ from the prior two in that pairs of
scenarios are presented in each experiment. One member of each pair is
as similar to the other member in as many financial aspects as possible.
They differ, however, in that only one member of each pair has a sunk
cost. In this way we can assess the impact of the sunk cost component
of the scenario.
Experiment 3
Question 3A. As the president of an airline company, you have invested 10
million dollars of the company’s money into a research project. The purpose was
to build a plane that would not be detected by conventional radar, in other words,
a radar-blank plane. When the project is 90% completed, another firm begins
marketing a plane that cannot be detected by radar. Also, it is apparent that their
plane is much faster and far more economical than the plane your company is
building. The question is: should you invest the last 10% of the research funds to
finish your radar-blank plane?
Yes 41
No 7
Question 3B. As president of an airline company, you have received a suggestion
from one of your employees. The suggestion is to use the last 1 million dollars of
your research funds to develop a plane that would not be detected by conventional
radar, in other words, a radar-blank plane. However, another firm has just begun
marketing a plane that cannot be detected by radar. Also, it is apparent that their
plane is much faster and far more economical than the plane your company could
build. The question is: should you invest the last million dollars of your research
funds to build the radar-blank plane proposed by your employee?
Yes 10
No 50
Question A vs B: x2(1, N = 108) = 50.6, p < .OOl.
The difference between these stories is that in question A millions have
already been invested, while in question B nothing has been invested yet.
Whereas the large majority of the question B respondents think the
project is a bad idea, question A respondents overwhelmingly endorse
continuing construction. There is no obvious economic reason to complete
the project. Yet there appears to be a compelling psychological one:
sunk cost.
While 3A respondents thought that continued spending was a much
better idea than 3B respondents did, we did not know if this would be
accompanied by an inflated certainty among 3A respondents that a completed
project would be a financial success. It may be that 3A respondents
grimly decide to spend despite desperate odds. On the other hand, perhaps
their willingness to throw good money after bad is due to the fact
that they do not perceive the situation as a lost cause. In Experiment 4
we sought the answer to this question.
130 ARKES AND BLUMER
Experiment 4
Questions 4A and 4B were identical to questions 3A and 3B. At the
end of each story subjects were told “Use the following 0 to 100 scale.
Write in the box the number between 0 and 100 that reflects what you
think your plane’s chance of financial success really is. You can use any
number.” The scale had five likelihood labels varying from “no chance”
at 0 to “sure thing” at 100.
The seventy-six 4A subjects’ mean probability estimate was 41.0. The
eighty-two 4B subjects’ mean probability estimate was 34.0, t(156) =
2.02, p < .05. We conclude that subjects in a sunk cost situation have an
inflated estimate of the likelihood that the completed project will be a
success. We do not know if this inflated estimate helps foster continued
investment, is a consequence of the decision to continue investing, or
both.
A possible flaw in Experiments 3 and 4 is that in questions 3A and 4A
the completed plane would be a 10 million dollar product. In questions
3B and 4B the plane would cost only 1 million. Perhaps 3B respondents
were unwilling to spend because they knew that a 1 million dollar plane
was likely to be a cheap, inferior product. Question A respondents would
not have such fears and therefore would be more likely to continue. In
Experiment 5 we tested this explanation of Experiments 3 and 4 by
changing the plane in question B to a 10 million dollar product.
Experiment 5
The story used in Experiment 5 was identical to that used in question
3B except that “1 million” was changed to “IO million.” Respondents’
decisions whether or not to build the plane were
Yes 10
No 50
Since the data from Experiment 5 were identical to those of 3B, we
conclude that the overwhelming decision not to build the radar-blank
plane in question 3B was not due to the smaller final price of the plane
compared to the price in 3A. We conclude that the critical difference in
questions 3A and 3B was that in the former the investor had incurred a
sunk cost.
EXPLANATION OF THE SUNK COST EFFECT
Thaler’s (1980) explanation of the sunk cost effect is based on prospect
theory (Kahneman & Tversky, 1979). Two features of prospect theory
seem particularly pertinent to the analysis of the sunk cost effect. The
first of the features is depicted in Fig. 1, which contains prospect theory’s
value function. This function represents the relation between objectively
SUNK COST 131
VALUE
+
LOSSES GAINS
FIG. I. The value function of prospect theory (Kahneman & Tversky, 1979).
defined gains and losses (e.g., measured in dollars) and the subjective
value a person places on such gains and losses.
When an initial investment is being considered, the investor is at point
A. After a substantial unsuccessful investment has been made, the
investor is at point B. At point B further losses do not result in large
decreases in value; however, comparable gains do result in large increases
in value. Therefore an investor at point B in Fig. 1 will risk small losses
in order to obtain possible large gains. Point B is the location of a person
who has paid a sunk cost. Compared to a person at point A, a person at
B is more likely to make a risky investment, i.e., to continue adding
funds to the sunk cost. This analysis is consistent with a finding by
McGlothlin (1956) that long shots at a race track are most popular during
the final race of the day. At that time many bettors are at point B and
are more risk prone than they were before any losses occurred.
Thaler (1980) has used prospect theory’s value function (Fig. 1) to
explain the snowstorm example presented earlier. The value of going to
the game is v(g). The value of losing $12 is 17(- 12), where 6 is the value
function for losses. The cost of enduring a snowstorm is c. We set the
enjoyment of the game equal to the cost of enduring the snowstorm, i.e.,
v(g) = - c( - c); therefore someone who received the tickets for free
would be indifferent about going to the game in a snowstorm. However,
the person who has already paid $12 will want to go since
v(g) + lq -(c + 12)) > 6( - 12).
The terms to the left of the “greater than” sign represent the net gain/
loss should the person go to the game. The term to the right of the
“greater than” sign represents the loss of $12-the result of not going
to the game. Call the ticket price p. Due to the convexity of Q, the second
term in the equation (c( - (c + p))) will always be smaller (i.e., closer to
132 ARKES AND BLUMER
zero) than (C(- c) + i;( - p)) for all p > 0. Thus the person who had paid
for the ticket will want to go.
A second feature of prospect theory pertinent to sunk costs is the
certainty effect. This effect is manifested in two ways. First, absolutely
certain gains 0, = 1.0) are greatly overvalued. By this we mean that the
value of certain gains is higher than what would be expected given an
analysis of a person’s values of gains having a probability less than 1.0.
Second, certain losses (p = 1.0) are greatly undervalued (i.e., further
from zero). The value is more negative than what would be expected
given an analysis of a person’s values of losses having a probability less
than 1.0. In other words, certainty magnifies both positive and negative
values.
Note that in question 3A the decision not to complete the plane results
in a certain loss of the amount already invested. Since prospect theory
states that certain losses are particularly aversive, we might predict that
subjects would find the other option comparatively attractive. This is in
fact what occurred. Whenever a sunk cost dilemma involves the choice
of a certain loss (stop the waterway project) versus a long shot (maybe
it will become profitable by the year 2500), the certainty effect favors the
latter option.
However, prospect theory does not specify the psychological basis
for the findings that sure losses are so aversive and sunk costs are so
difficult to ignore. One reason why people may wish to throw good money
after bad is that to stop investing would constitute an admission that the
prior money was wasted. The admission that one has wasted money
would seem to be an aversive event. The admission can be avoided by
continuing to act as if the prior spending was sensible, and a good way
to foster that belief would be to invest more. Staw (1976) showed that
when business school students felt responsible for a financially unsuccessful
prior decision, they continued to invest more money into that
option than if their prior decision was successful. This result seems to
contradict the commonsense notion that negative consequences will
cause a change in one’s course of action. Instead, Staw showed that
negative consequences fostered further commitment to the chosen alternative.
This unusual behavior is consistent with our own notion that sunk
costs are difficult to write off because to do so would appear wasteful.
We examined this proposed explanation in Experiment 6-another of
our relatively pure cases.
Experiment 6
On your way home you buy a tv dinner on sale for $3 at the local grocery store.
A few hours later you decide it is time for dinner, so you get ready to put the tv
dinner in the oven. Then you get an idea. You call up your friend to ask if he
SUNK COST 133
would like to come over for a quick tv dinner and then watch a good movie on tv.
Your friend says “Sure.” So you go out to buy a second tv dinner. However, all
the on-sale tv dinners are gone. You therefore have to spend $5 (the regular price)
for the tv dinner identical to the one you just bought for $3. You go home and put
both dinners in the oven. When the two dinners are fully cooked, you get a phone
call. Your friend is ill and cannot come. You are not hungry enough to eat both
dinners. You can not freeze one. You must eat one and discard the other. Which
one do you eat?
$3 dinner 2
$5 dinner 21
No preference 66
Since the costs and benefits of choosing either dinner are precisely
equal, we would expect, based on traditional economic theory, that everyone
would choose “no preference.” Sunk cost considerations, however,
heighten the attractiveness of the $5 dinner. Since the choice of the
$3 dinner is based on unknown factors, we deleted these respondents.
Of the remaining 87 subjects, 76% chose “no preference.” The 99% confidence
interval around this datum is 64 to 88%. We therefore conclude
that the prediction of traditional economic theory that 100% of the subjects
would have no preference is disconfirmed. We suggest that the
choice of the $5 dinner is made by many subjects because discarding it
would appear more wasteful than discarding the $3 dinner.
Another way to examine the role of wastefulness in the psychology of
sunk cost would be to write two stories that differ in that further expenditures
would appear to be wasteful in only one of the two stories. Subjects
should then demonstrate a greater willingness to spend in the other
story. For example, to buy a new printing press that is far better than
one’s present press would seem to be an excellent choice. But what if
one’s present press is rather new? To purchase another press-no matter
how superior- might then seem wasteful. Would the excellent choice
therefore be foresaken?
Experiment 7
Question 7A. As the owner of a printing company, you must choose whether to
modernize your operation by spending $200,000 on a new printing press or on a
fleet of new delivery trucks. You choose to buy the trucks, which can deliver your
products twice as fast as your old trucks at about the same cost as the old trucks.
One week after your purchase of the new trucks, one of your competitors goes
bankrupt. To get some cash in a hurry, he offers to sell you his computerized
printing press for $10,000. This press works 50% faster than your old press at
about one-half the cost. You know you will not be able to sell your old press to
raise this money, since it was built specifically for your needs and cannot be
modified. However, you do have $10,000 in savings. The question is should you
buy the computerized press from your bankrupt competitor?
Yes 49
No 15
134 ARKES AND BLUMER
Question 78. As the owner of a printing company, you must choose whether to
modernize your operation by spending $200,000 on a new printing press or on a
fleet of new delivery trucks. You choose to buy the press, which works twice as
fast as your old press at about the same cost as the old press. One week after your
purchase of the new press, one of your competitiors goes bankrupt. To get some
cash in a hurry, he offers to sell you his computerized printing press for $10,000.
This press works 50% faster than your new press at about one-half the cost. You
know you will not be able to sell your new press to raise this money, since it was
built specifically for your needs and cannot be modified. However, you do have
$10,000 in savings. The question is should you buy the computerized press from
your bankrupt competitor?
Yes 43
No 38
Question A vs B: x*(1, N = 145) = 7.51, p < .Ol.
Despite the fact that buying the printing press would result in the same
proportion of improvement in printing capability in questions 7A and 7B,
subjects in 7A were significantly more likely to buy the press. When
asked at the bottom of their questionnaires to give reasons for their
choice, the subjects in 7B who opted not to buy the press gave such
remarks as “I already have a good, new press that costs a lot of money.”
Subjects were less likely to buy if the purchase appeared to duplicate an
immediately prior one and therefore appeared to render it wasteful.4
Our explanation based on the appearance of wastefulness has an interesting
implication: if one’s own money is at stake or if one is personally
responsible for the initial investment (Staw, 1976), then wastefulness
should be more aversive than if someone else’s money is involved or if
someone else was responsible for the original investment decision. In
support of this conjecture Staw (1976) and Staw and Fox (1977) found
that personal responsibility for the situation did tend to increase further
financial allocations to a floundering investment.
In Experiment 8 we modified the questions used in Experiment 3 in
order to examine the role of personal involvement. In Experiment 3 you
are the president of an airline company. In Experiment 8 we described a
company in the third person.
Experiment 8
Question 8A. The Acme Airline Company has invested 10 million dollars of the
company’s money into a research project. The purpose was to build a plane that
would not be detected by conventional radar, in other words, a radar-blank plane.
4 A possible problem with Experiment 7 is that even though the purchase of a press would
result in an increase of SO% in printing capability in both stories, the purchase in 7A would
result in a slower machine than would the corresponding purchase in 7B. We ran another
experiment in which this difference in absolute printing speed was eliminated. The results
of this study replicated the results of Experiment 7.
SUNK COST 135
When the project is 90% completed, another firm begins marketing a plane that
cannot be detected by radar. Also, it is apparent that their plane is much faster
and more economical than the plane Acme is building. The question is should
Acme Airlines invest the last million dollars of its research funds to finish the
radar-blank plane?
Yes 37
No 21
Question 8B. The Acme Airlines Company has received a suggestion from one
of its employees. The suggestion is to use the last 1 million dollars of research
funds to build a plane that would not be detected by conventional radar, in other
words, a radar-blank plane. However, another firm has just begun marketing a
plane that cannot be detected by radar. Also, it is apparent that their plane is much
faster and far more economical than the plane your company could build. The
question is should you invest the last million dollars of your research funds to
build the radar-blank plane proposed by your employee?
Yes 2
No 35
Answers to question 8A differed significantly from those in 3A, x2(1,
N = 106) = 5.25, p < .OS. When you were the president of the company,
you were more likely to be a victim of the sunk cost effect than if you
were rendering judgment in general, Even when one’s general opinion
was solicited, however, the sunk cost effect was still very powerful (8A
vs 8B: x*(1, N = 95) = 29.5, p < .Ol).
In Experiment 9 we attempted to manipulate the personal involvement
present in Experiment 1. In the original experiment, you had paid for the
Michigan and Wisconsin ski trips. Would the sunk cost effect be diluted
if you had paid for neither trip?
Experiment 9
As you are listening to the radio one morning, the disk jockey from 95XIL’ calls
you. He informs you that you have won a free ski trip to either Michigan or
Wisconsin the last weekend of skiing season (which happens to be next weekend).
You think you will prefer the trip to Wisconsin rather than the trip to Michigan.
You call a travel agent and find out that the value of the Michigan ski trip is $100,
and the value of the Wisconsin ski trip is $50. You must decide which trip to take.
Which trip will you go on?
$100 ski trip to Michigan 44
$50 ski trip to Wisconsin 42
The x2 test comparing Experiment 1 (personal money at stake) to Experiment
9 (no personal money at stake) did not approach significance.
595XIL is a local radio station.
136 ARKES AND BLUMER
The analogous comparison between 8A and 3A had been significant. In
questionnaire studies such as these it is difficult to manipulate personal
commitment. Since in Experiments 8 and 9 we detected only equivocal
support for the hypothesis that personal involvement heightens the sunk
cost effect, we can draw no firm conclusions as yet on this point.
Experiment 10
We made one effort to ascertain whether economically sophisticated
subjects are less susceptible to the sunk cost effect. Fischhoff (1982) has
concluded that many attempts to debias people of their judgment errors
have not been successful. However, some of these debiasing efforts have
been of rather short duration, such as one experimental session. We
sought to determine if a college level economics course (or two) might
prove to be of value in lessening the sunk cost effect.
Method. One hundred twenty introductory psychology students were
divided into two groups based on whether or not they had ever taken a
college economics course. Fifty-nine students had taken at least one
course; sixty-one had taken no such course. All of these students were
administered the Experiment 1 questionnaire by a graduate student in
psychology. A third group comprised 61 students currently enrolled in an
economics course, who were administered the Experiment 1 questionnaire
by their economics professor during an economics class. Approximately
three fourths of the students in this group had also taken one prior
economics course. All of the economics students had been exposed to
the concept of sunk cost earlier that semester both in their textbook
(Gwartney & Stroup, 1982, p. 125) and in their class lectures.
Results. Table 1 contains the results. The x2 analysis does not approach
significance. Even when an economics teacher in an economics class
hands out a sunk cost questionnaire to economics students, there is no
more conformity to rational economic theory than in the other two
groups. We conclude that general instruction in economics does not
lessen the sunk cost effect.
TABLE 1
FREQUENCY OF CHOOSING EACH TRIP AS A FUNCTION OF ECONOMICS BACKGROUND
Student characteristic
$100 trip
$50 trip
Economics Psychology students
students with no economics
20 22
41 39
Psychology students
with economics
19
40
SUNK COST 137
RELATION TO OTHER THEORIES
Cognitive Dissonance Theory
The sunk cost effect would appear to be related to cognitive dissonance
theory (Festinger, 1957). Numerous studies have shown that once a subject
is induced to expend effort on an onerous task, the task is revalued
upward (e.g., Aronson & Mills, 1959). Such revaluation would presumably
result in increased willingness to expend further resources on the
task compared to the resources which would be voluntarily allocated by
a subject not having made a prior expenditure. This generally corresponds
to the sunk cost effect.
There are differences, however, between dissonance experiments and
the research presented here. First, in dissonance research subjects who
have no sufficient justification for performing an onerous task improve
their attitude toward the task. In the sunk cost situation, on the other
hand, it is unlikely that investors begin to enjoy their floundering investments.
Investors may pour good money after bad, but it is doubtful that
the bad money engenders positive feelings. Poor investments typically
engender substantial distress.
Second, cognitive dissonance theory would predict that a person who
endures suffering in order to attend the football game will enjoy the game
more than those who do not suffer. However, cognitive dissonance theory
does not predict who will be more motivated to attend the snowy game
(the contest winner or the ticket purchaser) unless we assume that the
act of paying for the ticket engendered suffering. This is most unlikely.
A more sensible assumption is that a person who has paid for the ticket
goes to the game because one knows that if one does not go, one will
later suffer. The reason suffering will later occur if one does not go to
the game is that one will feel regret for having wasted the admission price.
The cognitive dissonance analysis is mute on the question, “What causes
the later suffering,” which is answered directly by our wastefulness explanation.
We therefore feel that cognitive dissonance theory adds little
to our understanding of the sunk cost effect.
Entrapment
Another area that would seem to be relevant to the sunk cost effect is
the research on entrapment (e.g., Brockner, Shaw, & Rubin, 1979). Subjects
in entrapment situations typically incur small, continuous losses as
they seek or wait for an eventual goal. Brockner et al. cite the example
of waiting for a bus. After a very long wait, should you decide to take a
cab, thereby nullifying all the time you have spent waiting for the bus?
This is analogous to a sunk cost situation: time already spent waiting is
the sunk cost. Reluctance to call a cab represents the desire to invest
further in waiting.
138 ARKES AND BLUMER
In a recent analysis of entrapment experiments, Northcraft and Wolf
(1984) concluded that continued investment in many of them does not
necessarily represent an economically irrational behavior. For example,
continued waiting for the bus will increase the probability that one’s
waiting behavior will be rewarded. Therefore there is an eminently rational
basis for continued patience. Hence this situation is not a pure
demonstration of the sunk cost effect.
However, we believe that ~0rn.e sunk cost situations do correspond to
entrapment situations. The subjects who “owned” the airline company
would have endured continuing expenditures on the plane as they sought
the eventual goal of financial rescue. This corresponds to the Brockner
et al. entrapment situation. However, entrapment is irrelevant to the
analysis of all our other studies. For example, people who paid more
money last September for the season theater tickets are in no way
trapped. They do not incur small continuous losses as they seek an eventual
goal. Therefore we suggest that entrapment is relevant only to the
subset of sunk cost situations in which continuing losses are endured in
the hope of later rescue by a further investment.
Foor-in-the-Door and Low-Ball Techniques
Freedman and Fraser (1966) demonstrated that a person who first complies
with a small request is more likely to comply with a larger request
later. When the large and small requests are for related activities that
differ in their cost to the complying person, the phenomenon is called the
foot-in-the-door technique. An example would be first having people sign
a petition to encourage legislators to support safe driving laws. Later, the
petition signers are asked to display on their lawn a large sign that reads,
“Drive safely.”
When the small and large requests are for the Same target behavior,
the technique is called the “low-ball” procedure (Cialdini, Cacioppo,
Bassett, & Miller, 1978). An example would be getting someone to agree
to buy a car at a discounted price and then removing the discount. The
initial decision to buy heightens willingness to buy later when the car is
no longer a good deal.
Both the foot-in-the-door and low-ball techniques bear some similarity
to the sunk cost phenomenon in that an investment which is unlikely to
be made (question 3B) will be more likely if a prior commitment has
occurred (3A). A major difference between the sunk cost effect and the
two techniques is that compliance is the dependent variable with the two
techniques. Compliance typically plays no role in the sunk cost effect.
Most explanations of the foot-in-the-door technique are couched in
terms of self-perception theory (Bern, 1967). A person observes himself
or herself complying with a request to support good driving or a charitable
SUNK COST 139
organization. The person then concludes, “I’m the sort of person who
supports that cause.” This conclusion based on self-observation then
results in a high level of compliance when the larger request is made later
(Snyder & Cunningham, 1975).
We do not see how a self-perception explanation could readily be applied
to the sunk cost effect. We very much doubt that people continue
investing because they conclude they are the sort of person who continues
some particular investment. We suggest that the foot-in-the-door
technique applies largely to compliance and not the domain of the sunk
cost effect.
We believe that the low-ball procedure bears even less similarity to the
sunk cost effect. In the latter, a prior investment has occurred. In the
low-ball procedure, no prior investment has occurred, only a verbal commitment.
The buyer has not actually paid money to obtain the discounted
car. Therefore, in the sunk cost situation, an investment can be lost, while
in the low-ball procedure, there are no funds to be forfeited.
EPILOGUE
We have presented evidence which suggests that the sunk cost effect
is a robust judgment error. According to Thomas (1981), one person who
recognized it as an error was none other than Thomas A. Edison. In the
1880s Edison was not making much money on his great invention, the
electric lamp. The problem was that his manufacturing plant was not
operating at full capacity because he could not sell enough of his lamps.
He then got the idea to boost his plant’s production to full capacity and
sell each extra lamp below its total cost of production. His associates
thought this was an exceedingly poor idea, but Edison did it anyway. By
increasing his plant’s output, Edison would add only 2% to the cost of
production while increasing production 25%. Edison was able to do this
because so much of the manufacturing cost was sunk cost. It would be
present whether or not he manufactured more bulbs. Edison then sold
the large number of extra lamps in Europe for much more than the small
gBded manufacturing costs. Since production increases involved negligible
new costs but substantial new income, Edison was wise to increase
production. While Edison was able to place sunk costs in proper perspective
in arriving at his decision, our research suggests that most of
the rest of us find that very difficult to do.
REFERENCES
Aronson, E., & Mills, J. (1959). The effect of severity of initiation on liking for a group.
Journal of Abnormal and Social Psychology, 59, 177-181.
Bern, D. (1967). Self-perception: An alternative interpretation of cognitive dissonance phenomenon.
Psychological Review, 74, 183-200.
140 ARKES AND BLUMER
Brockner, J., Shaw, M. C., & Rubin, J. Z. (1979). Factors affecting withdrawal from an
escalating conflict: Quitting before it’s too late. Journal of Experimental Social Psychology,
15, 492-503.
Cialdini, R. B., Cacioppo, J. T., Bassett, R., & Miller, J. A. (1978). Low-ball procedure for
procuring compliance: Commitment then cost. Journal of Personality and Social Psychology,
36, 463-476.
Dowie, M. (1981). Atomic psyche-out. Mother Jones, 6, 21-23, 47-55.
Festinger, L. (1957). A theory of cognitive dissonance. Stanford: Stanford Univ. Press.
Fischhoff, B. (1982). Debiasing. In D. Kahneman, P. Slavic, & A. Tversky (Eds.), Judgment
under uncertainty: Heuristics and biases. New York: Cambridge Univ. Press.
Freedman, J. L., & Fraser, S. (1966). Compliance without pressure: The foot-in-the-door
technique. Journal of Personality and Social Psychology, 4, 195-202.
Grether, D. M., & Plott, C. R. (1979). Economic theory of choice and the preference reversal
phenomenon. American Economic Review, 69, 623-638.
Gwartney, J. D., & Stroup, R. (1982). Microeconomics: Private and public choice. New
York: Academic Press.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk.
Econometrica, 47, 263-291.
Lichtenstein, S. (1973). Response-induced reversals of preference in gambling: An extended
replication in Las Vegas. Journal of Experimental Psychology, 101, 16-20.
McGlothlin, W. H. (1956). Stability of choices among uncertain alternatives. American
Journal of Psychology, 69, 604-615.
Northcraft, G. B., & Wolf, G. (1984). Dollars, sense, and sunk costs: A life-cycle model of
resource-allocation decisions. Academy of Management Review, 9, 225-234.
Snyder, M., & Cunningham, M. R. (1975). To comply or not comply: Testing the selfperception
explanation of the “foot-in-the-door” phenomenon. Journal of Personality
and Social Psychology, 31, 64-67.
Staw, B. M. (1976). Knee-deep in the big muddy: A study of escalating commitment to a
chosen course of action. Organizational Behavior and Human Performance, 16,
27-44.
Staw, B. M., & Fox, E (1977). Escalation: Some determinants of commitment to a personally
chosen course of action. Human Relations, 30, 431-450.
Teger, A. I. (1980). Too much invested to quit. New York: Pergamon.
Thaler, R. (1980). Toward a positive theory of consumer choice. Journal of Economic
Behavior and Organization, 1, 39-60.
Thomas, R. P. (1981). Microeconomic applications: Understanding the American economy.
Belmont, CA: Wadsworth.
RECEIVED:October 10. 1983