Loss Aversion in Riskless Choice: A Reference-Dependent Model
Author(s): Amos Tversky and Daniel Kahneman
Source:
The Quarterly Journal of Economics,
Vol. 106, No. 4 (Nov., 1991), pp. 1039-1061
Published by: Oxford University Press
Stable URL: http://www.jstor.org/stable/2937956 .
Accessed: 26/04/2011 00:37
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .
http://www.jstor.org/action/showPublisher?publisherCode=oup. .
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Oxford University Press is collaborating with JSTOR to digitize, preserve and extend access to The Quarterly
Journal of Economics.
http://www.jstor.org
LOSS AVERSION IN
RISKLESS CHOICE:
A
REFERENCE-DEPENDENT MODEL*
AMos TVERSKY AND
DANIEL
KAHNEMAN
Much
experimental evidence indicates that
choice depends on the status quo or
reference level:
changes of reference point often
lead to reversals of preference. We
present
a
reference-dependent theory of consumer
choice, which explains such
effects
by a
deformation of indifference curves
about the reference point. The
central
assumption of the theory is that losses and
disadvantages have greater
impact on
preferences
than
gains
and
advantages.
Implications of loss aversion for
economic
behavior
are
considered.
The
standard models
of
decision
making assume that prefer-
ences
do
not
depend
on
current assets.
This assumption greatly
simplifies
the
analysis
of
individual choice
and
the
prediction
of
trades:
indifference
curves are
drawn without
reference
to
current
holdings, and the
Coase theorem asserts
that, except for transac-
tion
costs,
initial
entitlements
do not affect
final
allocations.
The
facts
of
the
matter are
more
complex.
There
is
substantial evidence
that
initial entitlements
do
matter and
that
the rate of exchange
between
goods
can
be quite different
depending
on
which is
acquired
and
which
is
given up,
even
in
the
absence
of
transaction
costs or
income
effects.
In
accord with a
psychological analysis
of
value,
reference levels play
a
large role
in
determining preferences.
In the present
paper we review the
evidence for this proposition
and offer a
theory
that
generalizes
the
standard model
by introduc-
ing a reference
state.
The
present analysis
of
riskless choice
extends
our
treatment
of
choice
under
uncertainty [Kahneman
and
Tversky, 1979, 1984;
Tversky
and
Kahneman, 1991], in which
the outcomes
of
risky
prospects
are
evaluated by
a
value function
that has three essential
characteristics.
Reference dependence:
the
carriers of
value are
gains
and
losses defined relative to a reference
point.
Loss
aversion:
the
function
is
steeper
in
the
negative
than
in
the
positive domain;
losses
loom
larger
than
corresponding
gains. Diminishing sensitiv-
ity:
the
marginal value
of
both
gains and
losses decreases with their
*This
paper has benefited from
the
comments of
Kenneth Arrow, Peter
Diamond,
David
Krantz,
Matthew
Rabin, and
Richard Zeckhauser. We are espe-
cially grateful
to
Shmuel
Sattath and
Peter
Wakker
for
their helpful suggestions.
This work
was
supported by
Grants
No. 89-0064
and
88-0206
from
the
Air
Force
Office of
Scientific
Research,
and
by
the Sloan
Foundation.
?
1991
by the President
and
Fellows
of
Harvard
College
and
the
Massachusetts Institute
of
Technology.
The
Quarterly Journal
of Economics, November
1991
1040
QUARTERLY
JOURNAL OF
ECONOMICS
size.
These
properties
give rise to an
asymmetric
S-shaped value
function,
concave above
the
reference point
and
convex
below it,
as
illustrated in
Figure
I.
In this
article
we apply
reference
dependence, loss
aversion,
and
diminishing
sensitivity
to the
analysis
of
riskless
choice.
To
motivate
this
analysis,
we
begin with a
review
of
selected
experimen-
tal
demonstrations.
I.
EMPIRICAL
EVIDENCE
The
examples
discussed in
this
section are
analyzed
by refer-
ence
to
Figure
II.
In
every
case
we
consider two
options
x
andy
that
differ
on
two valued
dimensions and show how
the
choice between
them is
affected
by
the
reference
point
from which
they
are
evaluated. The common
reason
for
these
reversals
of
preference
is
that the
relative
weight
of the
differences between x
and
y
on
dimensions
1
and 2
varies with the location of
the
reference value
on these
attributes.
Loss
aversion
implies
that the
impact
of a
difference on
a
dimension
is
generally greater
when
that
difference
is
evaluated
as a loss than
when
the same
difference is
evaluated
as
a
gain.
Diminishing
sensitivity implies
that the
impact
of
a
difference
is attenuated when both
options
are remote from
the
reference point for the
relevant
dimension. This
simple scheme
VALUE
LOSSES
GAINS
FIGURE
I
An
Illustration of
a
Value
Function
LOSS
AVERSION IN
RISKLESS
CHOICE
1041
serves
to
organize
a
large
set of
observations.
Although
isolated
findings
may be subject to
alternative
interpretations,
the
entire
body of
evidence provides
strong
support
for
the phenomenon of
loss aversion.
a. Instant
Endowment. An
immediate
consequence
of
loss
aversion is that
the loss
of
utility associated with
giving up
a
valued
good
is
greater than the
utility gain associated with
receiving
it.
Thaler
[1980]
labeled this
discrepancy the
endowment
effect,
because
value
appears
to
change when a
good
is
incorporated
into
one's endowment. Kahneman, Knetsch, and
Thaler
[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.
All
participants
received a
questionnaire.
The
form
given
to
the recipients
of
a
mug
(the
"sellers")
indicated that "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. 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
preferences 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.
As
shown
in
Figure II,
the choosers'
reference state
is
t,
and
they
face
a
positive
choice
between two
options
that
dominate
t;
receiving
a
mug
or
receiving
a sum
in
cash. The
sellers evaluate the same
options
from
y; they
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.
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
instan-
1042
Q
UAR TERLY JOURNAL OF ECONOMICS
S0o
0
ci)
E t
----
x
cz~~~~~~~~~~~r
o~~~~~~~~s
Dimension
1
FIGURE II
Multiple Reference
Points for the Choice Between x and
y
taneously, by giving
an individual
property rights
over a
consump-
tion
good.
The interpretation of the endowment effect may be illumi-
nated by
the
following thought experiment.
Imagine
that as
a
chooser
you prefer $4
over
a
mug.
You learn
that
most
sellers
prefer
the
mug
to
$6,
and
you
believe that
if
you
had
the
mug you
would
do
the
same.
In
light
of this
knowledge,
would
you
now
prefer
the
mug
over
$5?
If
you do, it is presumably because you have changed your
assessment
of the
pleasure
associated with
owning
the
mug.
If
you
still
prefer $4
over the
mug-which
we
regard
as
a
more
likely
response this indicates that you interpret
the
effect
of endow-
ment as
an
aversion
to
giving up your mug
rather than as an
unanticipated
increase
in
the
pleasure
of
owning
it.
b. Status
Quo
Bias. The retention of the status
quo
is an
option
in
many
decision
problems.
As illustrated
by
the
analysis
of
the sellers'
problem
in
the
example
of the
mugs,
loss
aversion
induces a bias that favors
the
retention
of
the status
quo
over other
options.
In
Figure II,
a decision maker who is indifferent between x
and
y
from
t
will
prefer
x
over
y
from
x,
and
y
over x
from y.
Sarnuelson
and Zeckhauser
[1988]
introduced the term "status
quo
bias"
for
this effect
of
reference
position.