Possession, feelings of ownership and the endowment effect

Judgment and Decision Making, Vol. 2, No. 2, April 2007, pp. 107–114
Possession, feelings of ownership and the endowment effect
Jochen Reb?
Terry Connolly
Singapore Management University
University of Arizona
Abstract
Research in judgment and decision making generally ignores the distinction between factual and subjective feelings
of ownership, tacitly assuming that the two correspond closely. The present research suggests that this assumption might
be usefully reexamined. In two experiments on the endowment effect we examine the role of subjective ownership by
independently manipulating factual ownership (i.e., what participants were told about ownership) and physical posses-
sion of an object. This allowed us to disentangle the effects of these two factors, which are typically confounded. We
found a signi?cant effect of possession, but not of factual ownership, on monetary valuation of the object. Moreover,
this effect was mediated by participants’ feelings of ownership, which were enhanced by the physical possession of the
object. Thus, the endowment effect did not rely on factual ownership per se but was the result of subjective feelings
of ownership induced by possession of the object. It is these feelings of ownership that appeared to lead individuals
to include the object into their endowment and to shift their reference point accordingly. Potential implications and
directions for future research are discussed.
Keywords: decision making; endowment effect; possession; psychological ownership, subjective ownership.
1 Introduction
of losing it is seen as a (relatively large) loss. If one
does not, the prospect of acquiring it is seen as a (rela-
Thaler (1980) presented half the students in a class with
tively small) gain. Hence the small volume of trading in
Cornell University coffee mugs and then allowed them
Thaler’s study: The endowment shifted reference points,
to trade with their less fortunate classmates. Surprisingly
and thus the assessment of what is a loss and what a gain.
little trading occurred. Those holding the mugs set their
More recently, researchers have started to examine in
minimum selling prices too high, and those without mugs
more detail the psychological mechanisms driving the ef-
set their maximum offers too low, for many trades to
fect. For example, Strahilevitz and Loewenstein (1998)
clear. Apparently, brie?y owning a coffee mug raised its
found that valuation of an object can increase with dura-
value to the owner suf?ciently to price it beyond the reach
tion of ownership, possibly due to increased adaptation,
of most non-owners. Thaler coined the term endowment
the psychological accustoming to the new material situ-
effect to describe the result: goods that are included in
ation. Novemsky and Kahneman (2005a, b) present ev-
one’s endowment — that is, goods that one owns — are
idence that loss aversion, and thus an endowment effect,
valued more highly than identical goods not held in the
is found for goods that are owned for consumption, but
endowment. The non-owner’s potential gain from acqui-
not for goods that are owned for exchange, and that are
sition was apparently smaller than the owner’s potential
thus given up “as intended” rather than as losses from an
loss from sale. The effect has since been widely repli-
endowment. Carmon and Ariely (2000) report ?ndings
cated (e.g., Kahneman, Knetsch, & Thaler, 1990; Kahne-
suggesting that endowment effects can be explained as
man, Knetsch, & Thaler, 1991).
the result of buyers and sellers having different cognitive
The endowment effect is commonly interpreted as the
perspectives on the exchange. These results suggest that
result of loss aversion, a core ingredient of prospect the-
there is more to the endowment effect than simple factual
ory (Kahneman & Tversky, 1979). Losses (outcomes
ownership of an object.
below some reference point) are weighted substantially
In the present research, we try to elucidate further what
more than gains (outcomes above the reference point)
leads to the development of a sense of endowment and,
in the evaluation of choice options (Kahneman & Tver-
subsequently, to higher monetary valuations. The con-
sky, 1984). If one initially owns an object, the prospect
cept of ownership of one’s endowment of goods appears
to involve two elements, legal entitlement and subjective
?We would like to thank Nidhi Chaudhry, Rashimah Binte Rajah,
ownership. These elements may be imperfectly corre-
Chintan Rastogi, Harsh Saxena, Malavika Shanker, Dionne Soriano,
lated. One may feel some sense of ownership of items
and Shangbin Xie for help with data collection and entry. Address:
one does not own (e.g., a borrowed bicycle) and behave
Jochen Reb, Lee Kong Chian School of Business, Singapore Man-
agement University, 50 Stamford Road, Singapore 178899. E-mail:
as an owner might (e.g., resenting the owner’s demand for
jreb@smu.edu.sg.
its return). Conversely, one may feel little ownership of
107

---

Judgment and Decision Making, Vol. 2, No. 2, April 2007
Possession, ownership and the endowment effect
108
items one does, in fact, own (e.g., a newly-bought com-
sign. The object to be evaluated was a chocolate bar of
puter) and require some period of experience and use be-
a brand familiar to the participants. We assessed mone-
fore feeling full ownership. Pierce, Kostova and Dirks
tary valuations as our dependent variable and feelings of
(2003) propose an elaborate psychological model, based
ownership as a potential mediator.
on an extensive literature review, of the antecedents, ex-
Participants in the Possession condition were given the
periences, and consequences of psychological or subjec-
chocolate bar before participating in an unrelated study
tive ownership (see Belk, 1988; Dittmar, 1992; Furby,
that took about 30 minutes. They were handed the choco-
1978). This research suggests that subjective feelings
late bar and told to keep it for now (and not to eat it) as
of ownership are more complex than simple legal enti-
it would be used later on in the study; they then placed
tlement (Etzioni, 1991). In the Pierce et al model feel-
it on the desk next to their computers. In the No Pos-
ings of ownership are induced by controlling the en-
session condition participants were shown the chocolate
tity (e.g., through possession), becoming familiar with it
bar only after participating in the unrelated study. They
(e.g., through actual or imagined use) and/or investing the
were not given possession of the item and they had no
self into it (e.g., through identi?cation).
physical contact with it. After completing the unrelated
In Thaler (1980), as in most studies of the endow-
study participants in the Ownership condition were told
ment effect, two of these elements of ownership are con-
that they now owned a chocolate bar (either the one on
founded. It seems clear that the students who received
their desk or one just like the one they had been shown)
the mugs understood that they legally owned them and
and that they could either keep it or exchange it for some
could, if they wished, sell them to others. They also were
money. Those in the No Ownership condition were told
given physical possession of the mugs and could inspect
that they now had a choice of receiving either a chocolate
and control them. The extent to which these elements
bar or some money. All participants then completed a re-
led to feelings of subjective ownership is unclear. Pos-
sponse sheet which presented a series of choices between
session alone, even in the absence of factual ownership,
the chocolate bar and amounts of money rising from $.10
can induce feelings of ownership (Etzioni, 1991; Furby,
to $4.00 in steps of 10 cents. Respondents indicated their
1980). Indeed, because of the immediate control it pro-
preference at each money amount. (All money amounts
vides over the entity, possession might be more psycho-
were given in the local currency, Singapore dollars, worth
logically salient and have a stronger effect on feelings of
at the time about .60 $US.)
ownership of an object (and thus on its monetary eval-
The experiment thus compares the preferences of the
uation) than does factual ownership (Pierce et al., 2003;
following groups: those who own a chocolate bar but
Rudmin & Berry, 1987).
may, if they wish, exchange it for money; and those who
In the following two experiments we examine the rel-
do not own a bar but are offered a choice of either ac-
ative contributions of factual and subjective ownership to
quiring one or receiving a sum of money.1 At the time of
the endowment effect by separately manipulating factual
making this choice half the participants had possessed the
ownership and possession of an object. This design al-
bar in the sense that they had received it from the experi-
lows examining whether the endowment effect is driven
menter and placed it on their desks for about 15–30 min-
by (a) factual ownership, (2) possession, or (3) both. We
utes. The other participants had not had this prior posses-
expect that possession will induce stronger feelings of
sion. As in previous studies (e.g., Strahilevitz & Loewen-
ownership than pure factual ownership as such. We are
stein, 1998) incentive compatibility was maintained by
arguing that it is this subjective sense of endowment —
telling the participants that the experimenter had previ-
rather than a legal entitlement — that leads to a shift in
ously written down an amount of money and that partic-
the reference point and makes not having the object feel
1
like a loss, rather than a foregone gain. As a result pos-
This method of assessing respondents’ valuation of an object from
their responses to a series of hypothetical choices between the item and
session will lead to higher monetary evaluation, whereas
differing sums of money has been used in a number of endowment ef-
factual ownership by itself will not.
fect studies (e.g., Kahneman et al., 1991; Lerner, Small, & Loewen-
stein, 2004) in which potential sellers (those initially given the object)
were compared not to potential buyers (those not given the object) but to
2 Experiment 1
“choosers”, who were offered choices such as those described above be-
tween the object and different sums of money. For current purposes, the
choice-based method has a number of advantages over other commonly-
2.1 Method
used measures such as Willingness to Pay (WTP) and Willingness to
Accept (WTA), which are prone to extreme responses (since they have
2.1.1 Design, procedure, manipulations, and mea-
no natural upper bound) and may be distorted if respondents interpret
them as opening bids in a bargaining session rather than as ?nal low-
sures
est or highest prices. Prices realized in actual market transactions may
be in?uenced by the particular market mechanism imposed (e.g., open
The experiment had a 2 (Ownership vs. No Ownership)
outcry, sealed bid, second price, etc.) and are, again, only imperfectly
x 2 (Possession vs. No Possession) between-subjects de-
related to individual valuations.

---

Judgment and Decision Making, Vol. 2, No. 2, April 2007
Possession, ownership and the endowment effect
109
ipants would receive either money or the chocolate bar
one who neither owns nor possesses it. When comparing
depending on the choice they had indicated on their re-
these two cells of our experiment we found a signi?cant
sponse sheet for this amount. Thus both under-bidders
effect, F(1, 47) = 4.83, p < .05, prep = .90, ?2 = .07. Those
and over-bidders faced the risk of receiving their less-
who were endowed with the object (i.e., owned and pos-
preferred option. Only honest valuations escape this con-
sessed it) gave higher monetary valuations, M = $1.79
cern.
(SD = .96), than those who were not endowed with the
All participants in a given session received the same
object (i.e., neither owned nor possessed it), M = $1.29
possession treatment to rule out possible social compari-
(SD = .55). Thus, the endowment effect was replicated.2
son effects between participants. Each session included
about 8 of the 99 participants, and sessions alternated
2.2.2 The relative roles of factual ownership and
between Possession and No Possession conditions. Par-
possession in the endowment effect
ticipants indicated their valuations on the response sheet
and then responded to additional questions including (1)
Was the endowment effect due to factual ownership of the
a measure of feelings of ownership (“How much do you
object, possession, or both? A 2 (Ownership) x 2 (Pos-
feel like you own the chocolate bar (even if you don’t
session) ANOVA showed a signi?cant main effect only
legally own it)?”) and (2) a measure of information about
for possession, F(1, 95) = 5.10, p < .05, prep = .92, ?2
the object (“How much information do you think you
= .04 (see Figure 1). Participants gave the chocolate bar
have in order to evaluate the chocolate bar?”), both on
a higher monetary value when they had possessed it (M
7-point scales. The experimenter then revealed his pre-
= $1.72, SD = .92) than when they had not possessed it
set valuation, and participants either kept or received a
(M = $1.35, SD = .66). The effect of actual ownership
chocolate bar or were given the amount of money (and
was not signi?cant (ownership, M = 1.60, SD = .88, no
had to return the chocolate bar), depending on the choice
ownership, M = 1.47, SD = .76), F(1, 95) = .64, ns, prep
they had indicated on their response sheets.
= .56, ?2 = .00, and there was no evidence of a signi?-
cant interaction, F(1, 95) = .00, ns, prep = .12, ?2 = .00.
2.1.2 Participants
These results suggest that the differences in valuation that
constitute the endowment effect were induced by posses-
Ninety-nine undergraduate students at a Singaporean uni-
sion, and subsequent feelings of ownership, rather than
versity participated as part of a larger experiment ses-
by ownership of the focal object as such.
sion in exchange for course credit and task payoffs as
described above.
Possession
2.0
Possession
No Possession
No Possession
2.1.3 Stimulus pretest
1.5
One might be concerned that the experimental manipula-
1.0
tions of ownership and possession are confounded with
the information participants received about the object,
Monetary valuation
0.5
which could affect monetary valuation. However, given
that a chocolate bar is a simple and familiar object we did
0.0
not anticipate any differences between experimental con-
Ownership
No Ownership
ditions in the amount of product information participants
Figure 1: Effects of factual ownership and possession on
had. Participants’ self-ratings con?rmed this. Judgments
monetary valuation, Experiment 1. (Error bars indicate
of whether they had suf?cient information to evaluate the
±1 standard error of the mean.)
chocolate bar (average rating of M = 4.16, (SD = 2.03),
on a 1–7 scale with higher values indicating more infor-
mation) showed no signi?cant difference between the two
possession conditions, F(1, 90) = 1.57, ns, p
2.2.3 The Role of Feelings of Ownership
rep = .72, ?2
= .01, or between the two ownership conditions, F(1, 90)
We next examined the possible mediating role of subjec-
= .40, ns, prep = .48, ?2 = .00.
tive feelings of ownership in linking possession to valua-
tion. A 2 (Ownership) x 2 (Possession) ANOVA revealed
2.2 Results and discussion
a signi?cant effect of possession on subjective ownership.
As predicted, participants who had the chocolate bar in
2.2.1 Replication of endowment effect
their possession felt stronger ownership (M = 4.34, SD
The typical endowment effect experiment compares an
2The selling-price/choice-price ratio of 1.39 is typical in magnitude
individual who both owns and possesses the object with
(e.g., in Lerner et al., 2004, it was 1.30).

---

Judgment and Decision Making, Vol. 2, No. 2, April 2007
Possession, ownership and the endowment effect
110
= 2.13) than those who did not (M = 2.60, SD = 1.86),
dure, design, and materials were identical to Experiment
F(1, 90) = 17.75, p < .001, prep = .999, ?2 = .15, (see Fig-
1 except for the differences noted below.
ure 2). As with valuations, we found no effect of factual
One purpose of Experiment 2 was to see if the results
ownership, F(1, 90) = .09, ns, prep = .30, ?2 = .00, and no
would replicate with a different object. For this exper-
signi?cant interaction F(1, 90) = .94, ns, prep = .62, ?2 =
iment, we chose the object that has probably been used
.00.
most commonly in endowment effect studies: a coffee
Feelings of ownership, in turn, predicted monetary
mug from the participants’ university. This object was
valuations. Regressing monetary valuations on feelings
rather different from the chocolate bar used in Experi-
of ownership showed that stronger feelings of owner-
ment 1, being more durable, non-hedonic, and of higher
ship were related to higher monetary valuations, B = .12,
price. To accommodate for the higher value, the new re-
SE(B) = .04, ? = .33, t(92) = 3.35, p < .01, adjusted R2
sponse sheet gave amounts of money rising from $.20 to
= .10. In order to test for mediation, we regressed par-
$8.00 in steps of 20 cents.
ticipants’ monetary valuations on both the experimental
In Experiment 1 the duration of possession and the du-
manipulation of possession and the presumed mediator
ration of ownership were unequal: participants in the Pos-
(feelings of ownership). Results showed that feelings of
session condition had the chocolate bar in their posses-
ownership continued to predict monetary valuation, ? =
sion for about 15–30 minutes, while those in the Own-
.29, t(90) = 2.64, p < .05, whereas the experimental ma-
ership condition only learned that they owned the item
nipulation did not, ? = .11, t(90) = .99, p = .32, ns. A
shortly before giving their valuations. This difference in
Sobel test of mediation was signi?cant, z = 2.24, p < .05.
duration may have exaggerated the effect of possession
Thus, we found that feelings of ownership fully mediated
relative to that of ownership, pitting “possession for a sig-
the effect of possessing the chocolate bar on the monetary
ni?cant period of time” against “ownership for a brief pe-
valuation of the item.
riod”. In Experiment 2, duration of possession and dura-
tion of ownership were equal (and short). As participants
7
sat down at individual cubicles for the experiment, they
were given a questionnaire explaining the study and the
6
condition they were in, and containing the valuation mea-
Possession
5
Possession
sures. All participants were shown an example mug, and
4
those in the Possession condition received the object at
No Possession
this time. After brief verbal instructions, all participants
3
No Possession
proceeded to give their valuations. The elapsed time from
Feelings of ownership
2
receiving experimental instructions to completing the val-
uation measures was thus only a minute or two, in both
1
Possession and Ownership conditions.
Ownership
No Ownership
Experiment 1 relied on a single dependent measure, the
Figure 2: Effects of factual ownership and possession on
monetary value at which the participant’s choice switched
feelings of ownership, Experiment 1.
from the object to a sum of money. In Experiment 2, we
added a second measure that asked participants to provide
In sum, Experiment 1 indicates that the sense of en-
willingness to accept (WTA) prices (for those in the Own-
dowment that leads to higher monetary valuations results
ership condition) and willingness to pay (WTP) prices
from the feelings of ownership induced by possessing an
(for those in the No Ownership condition). Participants
object, rather than legal ownership as such. Experiment 2
were asked “What is the minimum amount of money you
provides a conceptual replication, extending the ?rst ex-
are willing to accept for the coffee mug?” (WTA, Own-
periment to (a) a different object of choice, (b) an alterna-
ership condition), or “What is the maximum amount of
tive measure of the dependent measure, and (c) a differ-
money you are willing to pay for the coffee mug?” (WTP,
ent, and briefer, implementation of the concept of “pos-
No Ownership condition). We will refer to the ?rst set of
session.”
values as “choice values,” to the second as “WTA/WTP.”
In contrast to the choice values, which were re-
stricted by the response sheet to be between $0 and $8,
3 Experiment 2
WTA/WTP ratings were made in a free response format,
opening the possibility of extreme values that might un-
Experiment 2 was designed to provide a conceptual repli-
duly affect the analyses. Inspecting the responses re-
cation of the ?ndings of Experiment 1. The design was
vealed only one value that was markedly different ($20)
again a 2 (Ownership vs. No Ownership) x 2 (Possession
from the rest and double the next highest value ($10 for
vs. No Possession) between-subjects design. The proce-
two respondents). We Winsorized this value to the next

---

Judgment and Decision Making, Vol. 2, No. 2, April 2007
Possession, ownership and the endowment effect
111
highest ($10) (see Peters, Slovic, & Gregory, 2003 for a
F(1, 46) = 4.37, p < .05, prep = .89, ?2 = .06. Thus, the
similar approach).
endowment effect was again replicated.3
Finally, in Experiment 1 the Possession and No Posses-
sion conditions were conducted separately. This had the
3.2.2 The relative roles of factual ownership and
advantage of avoiding possible social comparison effects
possession in the endowment effect
between those who received the mug and those who did
not. However, this procedure had the disadvantage of not
Choice values. A 2 (Ownership) x 2 (Possession)
allowing for random assignment to one of the four exper-
ANOVA showed a signi?cant main effect only for pos-
imental cells. Thus, in Experiment 2, we ran all four cells
session, F(1, 91) = 5.13, p < .05, prep = .92, ?2 = .04
of the experiment at the same time, allowing for better
(see Figure 3). Participants gave the coffee mug a higher
random assignment of participants to experimental condi-
monetary value when they possessed it (M = $3.84, SD =
tions. We minimized potential social comparison effects
.2.03) than when they did not possess it (M = $3.00, SD
by seating participants in individual cubicles.
= 1.50). The effect of actual ownership was not signi?-
cant (ownership, M = 3.58, SD = 1.96, no ownership, M
= 3.30, SD = 1.73), F(1, 91) = .60, ns, p
3.1 Method
rep = .54, ?2 =
.00, and there was no evidence of a signi?cant interac-
3.1.1 Participants
tion, F(1, 91) = .77, ns, prep = .58, ?2 = .00. These results
replicate the ?ndings of Experiment 1 that the differences
Ninety-six business undergraduate students at a Singa-
in valuation that constitute the endowment effect were in-
porean university participated as part of a larger exper-
duced by possession rather than by factual ownership of
iment session in exchange for course credit and task pay-
the focal object.
offs as described in Experiment 1. Only Singaporean stu-
dents were allowed to participate as foreign exchange stu-
dents might perceive and value the university coffee mug
Possession
systematically different.
Possession
4
No Possession
No Possession
3.1.2 Stimulus Pretest
3
As in Experiment 1, we assessed participants’ judgments
2
of whether they had suf?cient information to evaluate the
coffee mug (average rating of M = 3.64, SD = 1.84, on
1
a 1–7 scale with higher values indicating more informa-
tion). These ratings showed no signi?cant difference be-
0
Monetary valuation (choice values)
tween the two possession conditions, F(1, 92) = 3.38, ns
Ownership
No Ownership
(p = .07), prep = .85, ?2 = .02, or between the two owner-
Figure 3: Effects of factual ownership and possession on
ship conditions, F(1, 92) = 1.23, ns, prep = .67, ?2 = .00.
choice values, Experiment 2.
3.2 Results and discussion
WTA/WTP. The same analysis on the WTA/WTP mea-
3.2.1 Replication of endowment effect
sure yielded the same substantive results: a main effect of
As in Experiment 1, we ?rst tested for a difference in val-
possession (possession, M = 4.01, SD = 2.29, no posses-
uation between (1) those who owned and possessed the
sion, M = 3.18, SD = 1.66), F(1, 92) = 4.10, p < .05, prep
object and (2) those who neither owned nor possessed it.
= .88, ?2 = .03. The effect of actual ownership was not
When comparing these two cells of our experiment, we
signi?cant (ownership, M = 3.85, SD = 2.18, no owner-
found a signi?cant effect on the choice measure, F(1, 46)
ship, M = 3.37, SD = 1.90), F(1, 92) = 1.45, ns, prep =
= 4.13, p < .05, p
.70, ?2 = .01, and there was no evidence of a signi?cant
rep = .88, ?2 = .06. Those endowed
with the object (i.e., owned and possessed it) gave higher
interaction, F(1, 92) = .69, ns, prep = .57, ?2 = .00.
monetary valuations, M = $3.82 (SD = 2.28), than those
not endowed (i.e., neither owned nor possessed it), M =
3.2.3 The Role of Feelings of Ownership
$2.70 (SD = 1.44). The same effect materialized with
the alternative, WTP/WTA dependent measure such that
We next examined again the possible mediating role of
those who owned and possessed the object (M = 4.08,
subjective feelings of ownership in linking possession to
SD = 2.61) gave higher valuations than those who nei-
3 The selling-price/choice-price ratio (1.41) is highly similar to that
ther owned nor possessed the it (M = 2.76, SD = 1.64),
of Experiment 1 (1.39).

---

Judgment and Decision Making, Vol. 2, No. 2, April 2007
Possession, ownership and the endowment effect
112
Possession
Possession
No Possession
4 General Discussion
4
No Possession
3
In two endowment effect experiments, we independently
manipulated factual ownership and possession of an ob-
2
ject (a chocolate bar in Experiment 1 and a university cof-
fee mug in Experiment 2). This allowed us to disentangle
1
the effects of these two factors, which are typically con-
Monetary valuation (WTA/WTP)
0
founded — sellers generally both own and possess the
Ownership
No Ownership
object from the outset, while buyers or choosers neither
own nor possess it. We found a signi?cant effect of pos-
Figure 4: Effects of Factual Ownership and Possession
session, but no signi?cant effect of factual ownership, on
on WTA/WTP, Experiment 2
monetary valuation of the objects, both on choice values
(Experiment 1 and 2) and traditional WTA/WTP values
(Experiment 2). These results suggest that while the en-
valuation. A 2 (Ownership) x 2 (Possession) ANOVA
dowment effect can be “turned on” in the typical fashion,
revealed a signi?cant effect of possession on subjective
it can be “turned off” in one of two ways. First, it can be
ownership. As predicted, participants who had the cof-
turned off by taking the good away from both sellers and
fee mug in their possession felt stronger ownership (M =
buyers (no possession). Second, it can be given to both
3.54, SD = 1.79) than those who did not (M = 2.48, SD
sellers and buyers (possession). These results help to clar-
= 1.48), F(1, 92) = 9.94, p < .01, prep = .98, ?2 = .09,
ify the antecedents of the endowment effect and suggest
(see Figure 4). As with valuations, we found no effect of
an important role of the subjective sense of ownership in
factual ownership, F(1, 92) = .63, ns, prep = .55, ?2 = .00,
decision making.
and no signi?cant interaction F(1, 92) = .08, ns, prep =
The results suggest that the endowment effect may
.29, ?2 = .00.
be primarily driven by subjective feelings of ownership
Stronger feelings of ownership, in turn, led to higher
rather than by factual ownership as such. In other words,
monetary valuations, B = .46, SE(B) = .10, ? = .43, t(93)
it may require the development of a subjective sense of
= 4.59, p < .001, adjusted R2 = .18. In order to test for me-
endowment, rather than a legal entitlement, for the ref-
diation, we regressed participants’ choice values on both
erence point to shift. Once the reference point is shifted,
the experimental manipulation of possession and the pre-
loss aversion sets in and leads to higher valuations. In our
sumed mediator (feelings of ownership). Results showed
experiments, this shift seems to have been triggered by
that feelings of ownership continued to predict monetary
possession, not factual ownership. Thus, it appears that
valuation, ? = .40, t(92) = 4.01, p < .001, whereas the
participants in the possession condition felt that not hav-
experimental manipulation did not, ? = .11, t(92) = 1.06,
ing the object would be a loss, rather than a foregone gain
p = .29, ns. A Sobel test of mediation was signi?cant, z =
— counterfactually to their objective state of no owner-
2.49, p < .05. Thus, we found that feelings of ownership
ship in the possession/no-ownership condition. On the
fully mediated the effect of possessing the coffee mug on
other side, participants in the no possession condition felt
the monetary valuation of the item.4
that not having the object would be a foregone gain, rather
7
than a loss — again counterfactually to their objective
6
state of ownership in the no-possession/ownership condi-
tion.
5
Possession
This interpretation is consistent with the results of the
4
Possession
mediation analyses. In both experiments, we found the
3
No Possession
No Possession
effect of possession on monetary valuation to be com-
pletely mediated by participants’ rated feelings of own-
Feelings of ownership
2
ership, which were enhanced by the physical posses-
1
sion of the object (brie?y in Experiment 2, for a sig-
Ownership
No Ownership
ni?cant period of time in Experiment 1). The results
of these mediation analyses should be interpreted cau-
Figure 5: Effects of factual ownership and possession on
tiously, however, due the inherent shortcomings in obser-
feelings of ownership, Experiment 2.
vational/correlational approaches such as mediation anal-
yses (Spencer, Zanna, & Fong, 2005) and because of
a potential in?uence of the monetary valuations partici-
4The same substantive results were obtained for WTA/WTP as de-
pendent measures. For reasons of brevity, the analyses are omitted, but
pants provided before indicating their subjective feelings
can be obtained from the ?rst author.
of ownership.

---

Judgment and Decision Making, Vol. 2, No. 2, April 2007
Possession, ownership and the endowment effect
113
The present research contributes to a growing liter-
Journal of Consumer Research, 15, 139–168.
ature on the psychological mechanisms behind the en-
Carmon, Z., & Ariely, D. (2000). Focusing on the for-
dowment effect (e.g., Carmon & Ariely, 2000; Novem-
gone: How value can appear so different to buyers and
sky & Kahneman, 2005a,b) and may help explain several
sellers. Journal of Consumer Research, 27, 360–370.
past ?ndings. For example, Strahilevitz and Loewenstein
Carmon, Z., Wertenbroch, K., & Zeelenberg, M. (2003).
(1998) found that valuation of an object increased with
Pre-choice attachment to the alternatives:
When
the duration of ownership. It is possible that over time,
choosing feels like losing. Journal of Consumer Re-
owners’ feelings of ownership increased, thus leading to
search, 30, 15–29.
higher valuations. Lerner, Small, & Loewenstein (2004)
Dittmar, H. (1992). The social psychology of material
found that owners’ selling prices of an object decreased
possessions. New York: St. Martin’s Press.
strongly when incidental disgust was induced experimen-
Etzioni, A. (1991). The socio-economics of property.
tally (through a video clip). It is possible that the experi-
Journal of Social Behavior and Personality, 6, 465–
ence of disgust, which is associated with an “expel” goal
468.
(see Lerner et al.), prevented the development of feelings
Furby, L. (1978). Possessions: Toward a theory of their
of ownership for the object, thus leading to lower mone-
meaning and function throughout the life cycle. In P.
tary valuations as compared to a control group.
B. Baltes (Ed.), Life span development and behavior
Beyond the endowment effect, the present research
(pp. 297–336). New York: Academic Press.
also raises the possibility that the notion of subjective
Furby, L. (1980). The origins and early development of
ownership may help explain a number of phenomena that
possessive behavior. Political Psychology, 2, 30–42.
have been of interest to decision researchers. For exam-
Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1990).
ple, future research could examine the role of subjec-
Experimental tests of the endowment effect and the
tive ownership in the escalation of commitment to fail-
Coase Theorem. Journal of Political Economy, 98,
ing projects (Staw, 1976) and failure to disregard sunk
1325–1348.
costs (Arkes & Blumer, 1985), which may well involve
Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1991).
decision makers’ subjective ownership of a project and
Anomalies: The endowment effect, loss aversion, and
a resulting feeling that one must take care of and main-
status quo bias. Journal of Economic Perspectives, 5,
tain it. Similarly, late-bid escalation in English auc-
193–206.
tions (Ku, Malhotra, & Murnighan, 2005) may be partly
Kahneman, D., & Tversky, A. (1979). Prospect theory:
due to strong feelings of ownership developing among
An analysis of decision under risk. Econometrica, 47,
the few remaining bidders. Subjective ownership might
263–291.
also help explain why windfall gains (Arkes et al., 1994)
Kahneman, D., & Tversky, A. (1984). Choices, values,
and “house money” (Thaler & Johnson, 1990) are spent
and frames. American Psychologist, 39, 341–350.
lightly, and why pre-choice attachment to more than one
Ku, G., Malhotra, D., & Murnighan, J. K. (2005).
option can lead to post-choice discomfort after having
Towards a competitive arousal model of decision-
to choose one of them, which implies “losing” the other
making: A study of auction fever in live and Internet
(Carmon, Wertenbroch, & Zeelenberg, 2003). Thus, a
auctions. Organizational Behavior and Human Deci-
more nuanced conceptualization of the possible diver-
sion Processes, 96, 89–103.
gence of objective and subjective ownership may help to
Lerner, J. S., Small, D. A., Loewenstein, G. (2004). Heart
explain a number of important decision making phenom-
strings and purse strings: Carryover effects of emo-
ena. As usual, more research is needed before we can
tions on economic decisions. Psychological Science,
fully appreciate the role of feelings of ownership in judg-
15, 337–341.
ment and decision making.
Novemsky, N., & Kahneman, D. (2005a). The bound-
aries of loss aversion. Journal of Marketing Research,
42, 119–128.
References
Novemsky, N., & Kahneman, D. (2005b). How do inten-
tions affect loss aversion? Journal of Marketing Re-
Arkes, H. R., & Blumer, C. (1985). The psychology of
search, 42, 139–140.
sunk cost. Organizational Behavior and Human Deci-
Pierce, J. L., Kostova, T., & Dirks, K. T. (2003). The state
sion Processes, 35, 124–140.
of psychological ownership: Integrating and extending
Arkes, H. R., Joyner, C. A., Pezzo, M. V., Gradwohl
a century of research. Review of General Psychology,
Nash, J., Siegel-Jacobs, K., & Stone, E. (1994). The
7, 84–107.
psychology of windfall gains. Organizational Behav-
Rudmin, F. W., & Berry, J. W. (1987). Semantics of own-
ior and Human Decision Processes, 59, 331–347.
ership: A free-recall study of property. Psychological
Belk, R. W. (1988). Possessions and the extended self.
Record, 37, 257–268.

---

Judgment and Decision Making, Vol. 2, No. 2, April 2007
Possession, ownership and the endowment effect
114
Peters, E., Slovic, P., & Gregory, R. (2003). The role of
Strahilevitz, M. A., & Loewenstein, G. (1998). The ef-
affect in the WTA/WTP disparity. Journal of Behav-
fect of ownership history on the valuation of objects.
ioral Decision Making, 16, 309–330.
Journal of Consumer Research, 25, 276–289.
Spencer, S. J., Zanna, M. P., & Fong, G. T. (2005). Es-
Thaler, R. H. (1980). Toward a positive theory of con-
tablishing a causal chain: Why experiments are often
sumer choice. Journal of Economic Behavior and Or-
more effective than mediational analyses in examining
ganization, 1, 39–60.
psychological processes. Journal of Personality and
Thaler, R. H., & Johnson, E. J. (1990). Gambling with the
Social Psychology, 89, 845–851.
house money and trying to break even: The effects of
Staw, B. M. (1976). Knee-deep in the big muddy: A
prior outcomes on risky choice. Management Science,
study of escalating commitment to a chosen course of
36, 643–660.
action. Organizational Behavior and Human Perfor-
mance, 16, 27–44.

---