Richard A. Easterlin - Specialized Research AI at MIT

Richard A. Easterlin

The economics of happiness

Most of us, I think it is safe to say,

would like to be happier, would like,
indeed, to hold the ‘keys to happiness.’
For centuries the contemplation of this
desire was the exclusive preserve of phi-
losophers and theologians, who specu-
lated and offered prescriptions on ‘the
good life.’ Only fairly recently has it
come into the domain of social science–
½rst in psychiatry, where depression had
been the object of concern, and then,
since around 1950, in the mainstream so-
cial sciences. The impetus for social sci-
ence research in this area during the last
half century has been the development
of population surveys inquiring into
people’s feelings of well-being. A very
simple survey question, for example,
might ask a respondent, “In general,

Richard A. Easterlin, a Fellow of the American
Academy since 1978, is professor of economics at
the University of Southern California. He is past
president of the Population Association of Ameri-
ca and of the Economic History Association. His
writings include “Population, Labor Force, and
Long Swings in Economic Growth: The American
Experience” (1968), “Birth and Fortune: The Im-
pact of Numbers on Personal Welfare” (1980),
and “Growth Triumphant: The 21st Century in
Historical Perspective” (1996).

how happy would you say you are–very
happy, pretty happy, or not so happy?”
Another question might be, “How satis-
½ed are you with your life as a whole–
very, somewhat, so-so, not very, or not at
all?”

Over the years a substantial method-
ological literature has developed to con-
sider the value of the answers to such
questions. The professional consensus is
that the responses, though not unprob-
lematic, are meaningful and reasonably
comparable among various groups of in-
dividuals. Although there are subtle dif-
ferences between happiness and life sat-
isfaction, I will treat them for the present
purpose as interchangeable measures of
overall feelings of well-being, that is,
of subjective well-being. My focus will be
on what we are learning from the survey
data on the causes of subjective well-
being, and, based on this, what we might
do as individuals to improve it.

As I go along I shall discuss two promi-

nent and contrasting theories of well-
being, one from psychology, one from
economics. In psychology, setpoint theo-
ry has gained increasing attention in
the last decade or so. Each individual is
thought to have a ½xed setpoint of hap-
piness or life satisfaction determined by
genetics and personality; life events such
as marriage or divorce, job loss, or seri-
ous injury or disease may temporarily

Dædalus Spring 2004

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
3
2
2
6
1
8
2
8
7
8
8
0
0
1
1
5
2
6
0
4
3
2
3
0
4
9
3
6
1
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

deflect a person above or below this set-
point, but in time each individual will
adjust to the new circumstances and re-
turn to the given setpoint. Psychologists
call this adjustment process ‘hedonic
adaptation.’ One setpoint theory writer
states flatly that life circumstances have
a negligible role to play in a theory of
happiness. If this is correct, then there is
little that you or I can do to improve our
well-being, and public policies aimed at
making people better off by improving
their social and economic conditions are
fruitless.

In contrast, economics places particu-

lar stress on the importance of life cir-
cumstances–particularly on one’s in-
come and employment situation–to
well-being. The view that money makes
you happier ½nds ringing endorsement
in economic theory. The implication is
that one can improve one’s happiness by
getting more money, and that public pol-
icy measures aimed at increasing the in-
come of society as a whole will increase
well-being.

I shall argue that the accumulating sur-

vey evidence indicates that neither of
these theories is correct. Contrary to
setpoint theory, life events such as mar-
riage, divorce, and serious disability or
disease do have lasting effects on happi-
ness. Contrary to economic theory,
more money does not make people hap-
pier. My discussion will be guided here
by what people themselves say about
what makes them happy.

In the early 1960s, social psychologist

Hadley Cantril carried out an intensive
worldwide survey in fourteen countries,
rich and poor, capitalist and communist,
asking open-ended questions about
what people want out of life–what they
would need for their lives to be com-
pletely happy. I would like to stress the
open-ended nature of Cantril’s survey.

There have been many surveys of peo-
ple’s values and goals, but almost all
present the respondent with a list prede-
termined by the interviewer. Cantril, in
contrast, let each respondent speak for
him or herself.

Despite the enormous socioeconomic
and cultural disparities among the coun-
tries, what people say in Cantril’s survey
is strikingly similar. In every country,
material circumstances, especially ma-
terial living conditions, are mentioned
most often. Next are family concerns
such as a happy family life. This is fol-
lowed by concerns about one’s personal
or family health. After this, and about
equal in importance, are matters related
to one’s work (e.g., having an interesting
job) and to personal character (emotion-
al stability, personal worth, self-disci-
pline, etc.). Concerns about broad inter-
national or domestic issues, such as war,
political or civil liberty, and social equal-
ity, are rarely mentioned.

Thus, it is the things that occupy most

people’s everyday lives, and are some-
what within their control, that are typi-
cally at the forefront of their personal
concerns–especially making a living,
marriage and family, and health. The
universality of these concerns helps
explain why comparisons of happiness
among groups of individuals are mean-
ingful: most people base their judgments
of well-being on essentially the same
considerations.

In what follows I shall discuss the
evidence on the relation between happi-
ness and the three circumstances people
most often name as their sources of well-
being–material living conditions, fami-
ly circumstances, and health. I will focus
throughout on average relationships.
Needless to say, what is true on average
is not necessarily true for each individ-
ual; but it is important to be clear on
what is typical. I’ll be reporting the re-

The
economics
of happiness

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
3
2
2
6
1
8
2
8
7
8
8
0
0
1
1
5
2
6
0
4
3
2
3
0
4
9
3
6
1
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Dædalus Spring 2004

Richard A.
Easterlin
on
happiness

sults of survey data–some, but not all,
from my own research–that show how
life events affect well-being as people
progress through the adult life cycle,
from early adulthood through middle
age to retirement.

Most of the generalizations in the
social science literature on subjective
well-being are based not on life-cycle
but on point-of-time studies. As shall be
seen below in regard to money and hap-
piness, point-of-time relationships are
not always replicated over the life
course. Even in those studies that do try
to follow the same individuals over time,
the period covered is rarely more than a
year or two; hardly ever are data repre-
sentative of the national population as a
whole available for as long as ½ve or ten
years. The life-cycle approach that I use
here employs the demographers’ tech-
nique of birth-cohort analysis and cov-
ers a much longer segment of the life
course. Although the same individuals
are not interviewed in each successive
year, we do have a nationally representa-
tive random sample of the same group
of individuals–of those born in a given
decade. The special advantage of this
approach is that we can follow birth co-
horts in American data on happiness for
almost thirty years.

Let me start with health. The critical

issue is whether signi½cant changes in
health have a lasting effect on happiness.
One might suppose, on the one hand,
that a serious accident or major disease
would permanently reduce one’s happi-
ness. On the other hand, people may
bounce back from such occurrences,
especially if helped by medications and
health devices such as wheelchairs, and
by a support network of friends and rela-
tives.

Indeed, the psychologists’ setpoint
theory sees people as adapting fully and

returning to the level of happiness that
they had before the adverse turn in
health. The seminal article, repeatedly
cited in the psychological literature as
evidence of complete adaptation, is a
1978 study of twenty-nine paraplegics
and quadriplegics by psychologist Phil-
ip Brickman and his collaborators. The
study’s principal conclusion is that the
accident victims, when compared with
twenty-two people who were compara-
ble in all respects except that they had
not experienced serious disability, “did
not appear nearly as unhappy as might
have been expected.” As a careful read-
ing of this statement makes clear, the
study does not actually assert that there
was complete adaptation. Indeed, the
statistical ½nding is that the accident
victims were signi½cantly less happy
than the comparison group.

There have been a number of studies
since, some continuing to claim com-
plete adaptation, others contradicting it.
To my knowledge the most comprehen-
sive investigation is a 1990 American in-
quiry that compares the life satisfaction
of large national samples of disabled and
nondisabled persons. The conclusion is
that the life satisfaction of those with
disabilities is, on average, signi½cantly
lower than that of those who report no
disabilities. Even more telling is the ½nd-
ing that when persons with disabilities
are classi½ed in several different ways–
according to the severity of the disabili-
ty, to whether the respondent suffers
from one or multiple conditions, to what
extent the respondent is limited in daily
activities, and to whether close contacts
are thought to perceive the respondent
as disabled–life satisfaction is lower for
those with more serious problems on
every single one of these dimensions.

It is highly unlikely that these system-
atic differences in life satisfaction arise
because those with worse problems sim-

Dædalus Spring 2004

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
3
2
2
6
1
8
2
8
7
8
8
0
0
1
1
5
2
6
0
4
3
2
3
0
4
9
3
6
1
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

ply haven’t had enough time to adapt.
The more straightforward conclusion is
that, on average, an adverse change in
health permanently reduces happiness,
and the worse the change in health, the
greater the reduction in happiness. The
results do not mean that no adaptation
to disability occurs. But the evidence
does suggest that even with adaptation,
there is, on average, a lasting negative
effect of poor health on happiness.

Let me turn from this point-in-time
evidence to some relating to the life cy-
cle. As we all know, among adults real
health problems increase as people
age. But what do people say about their
health? If people adapted completely to
adverse changes in health, as setpoint
theory asserts, then there should be no
change in self-reported health over the
life course because people would contin-
uously adjust to worsening health. Is it
true that self-reported health doesn’t
change?

The answer is no; self-reported health

declines throughout the life course. If
one follows Americans born in the de-
cade of the 1950s over the twenty-eight-
year segment of the life span for which
data are available, one ½nds a clear and
statistically signi½cant downtrend in
their average self-reported health. This
downtrend in self-reported health as
people get older is also true of people
born in earlier decades as far back as
reliable data go.

This ½nding assesses adaptation in
terms of self-reported health–not life
satisfaction, as in the disability analysis.
However, it is not the case that people
with worsening health do not feel un-
happy about it: they say that they are
less satis½ed with their health and that
they are less happy generally. At a point
in time among adults of all ages, report-
ed happiness is always less, on average,
the poorer the state of self-reported

health. This can be due in part to a loss
of income, but more importantly to non-
pecuniary effects such as limits on one’s
usual activities. It seems clear from com-
prehensive survey evidence that, con-
trary to the psychologists’ setpoint theo-
ry, adverse health changes have a lasting
negative effect on happiness, and that
there is less than complete adaptation to
deteriorating health.

Let me now turn to the effects of mar-

riage and marital dissolution. One might
suppose that establishing close and inti-
mate relationships of the sort marriage
embodies would typically make the part-
ners in such relationships happier and
more satis½ed with life in general. Some
of the initial pleasure of a new union
would be expected to wear off in time;
similarly, people who have lost a part-
ner through death, separation, or di-
vorce would be expected to adjust some-
what to single status. But, on average,
the close relationships embodied in mar-
riage would be expected to have a lasting
positive effect on one’s happiness, and
the loss of such relationships, a perma-
nently negative effect. (I am using ‘mar-
riage’ here as a proxy for the formation
of unions. These days marriage is some-
times preceded by a period of cohabita-
tion, and the real union consequently
takes place some time prior to mar-
riage.)

The psychologists’ setpoint theory
would argue, however, that adaptation
to marriage and marital dissolution is
complete. Indeed, there is a recent em-
pirical study of the German population
claiming to support this conclusion.
This study holds that around the time
of marriage, happiness increases briefly
during what is proverbially called the
honeymoon period, but that after one
year it returns to the level that prevailed
more than one year before marriage.

The
economics
of happiness

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
3
2
2
6
1
8
2
8
7
8
8
0
0
1
1
5
2
6
0
4
3
2
3
0
4
9
3
6
1
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Dædalus Spring 2004

Richard A.
Easterlin
on
happiness

Meanwhile, widowhood takes a some-
what longer time–eight years–for com-
plete adaptation to occur. (Separation
and divorce were not included in the
study.)

American data, however, suggest that
the formation and dissolution of unions
produce enduring effects, thus contra-
dicting the results of the German study.
As the proportion of married Americans
between the ages of eighteen and twen-
ty-nine increases, the average happiness
of those who marry is consistently high-
er than that of the unmarried, and quite
constant.

If these young Americans were simply

experiencing a temporary increase in
well-being when they married, their
average happiness should peak at ages
eighteen to nineteen, when all or almost
all of those married are in the honey-
moon period. Thereafter, average happi-
ness should progressively decline as it
returns to the setpoint level for an ever-
larger proportion of those married. But
in fact, throughout the ½rst decade of
marriage the happiness of young mar-
ried persons remains constant at a high-
er level than that of their unmarried
counterparts.

The American results also contradict
the argument that the higher happiness
of the married group stems from a selec-
tion effect–that those getting married
are happier to start with. If they are hap-
pier to start with, then the life satisfac-
tion of the combined group of married
and unmarried people would not in-
crease as more and more people mar-
ry. But the happiness of the group as a
whole, married and unmarried, does in-
deed increase as the proportion of mar-
ried people between the ages of eighteen
and twenty-nine rises.

The survey evidence continues to sug-
gest that there are lasting effects associ-
ated with marital status beyond the early

adult years. The happiness of married
people remains signi½cantly greater than
that of the unmarried throughout the
life cycle. People who remarry are just
as happy as those still in their ½rst mar-
riage; and even after thirty-½ve years
of marriage, the happiness of those still
in their ½rst marriage continues to be
signi½cantly greater than that of their
unmarried counterparts.

Results consistent with these are re-
ported by American sociologist Linda
Waite and her collaborators in a nation-
ally representative study that follows
½ve thousand married Americans over a
½ve-year period. At the end of the peri-
od, the happiness of those still married
is virtually unchanged, while the happi-
ness of those who separated, or divorced
and did not remarry, is signi½cantly low-
er. Remarriage reverses the effect of di-
vorce–those who divorced and then
remarried experience about the same
level of happiness as those who stayed
married. The lesson is clear: on average,
marriage brings greater happiness, mari-
tal dissolution, less.

Evidence of people’s desires for a ‘hap-
py marriage’ also contradicts the notion
that people adapt completely to their
marital circumstances. Six in ten people
cite a happy marriage as a factor when
asked about their conception of the
good life. More remarkable is how
women over forty-½ve who have never
married answer this question. Among
these women more than four in ten cite
a happy marriage as part of the good life
as far as they personally are concerned.
Perhaps some have adapted, and doubt-
less some never wanted to marry in the
½rst place–but a sizable proportion of
these women who have been single their
entire lives has not fully adjusted to the
possibility of never marrying.

These are substantial reasons, I be-
lieve, for concluding that adaptation

Dædalus Spring 2004

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
3
2
2
6
1
8
2
8
7
8
8
0
0
1
1
5
2
6
0
4
3
2
3
0
4
9
3
6
1
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

with regard to marital status is less than
complete, that the formation of unions
has a lasting positive effect on happi-
ness, and that dissolution has a perma-
nently negative effect. If the psychol-
ogists’ setpoint model is correct that life
circumstances are of negligible impor-
tance to long-run happiness, then it is
hard to see how one can reconcile it with
the bulk of population survey evidence
on either marriage or health.

Let me briefly mention two other
pieces of survey evidence that are dif½-
cult to square with the setpoint model.
First, throughout the life cycle, blacks
in the United States are, on average, con-
sistently less happy than whites. One
would be hard put, I believe, to argue
that this difference is due simply to dif-
ferent setpoints given by genetics and
personality, and that differences in the
life circumstances of the two races are
of little importance. Second, beyond age
sixty the life-cycle excess of female over
male happiness is reversed. Clearly, this
cannot be explained by genetic and per-
sonality factors; rather an important life
event–the much higher incidence of
widowed women than widowed men–
is chiefly responsible.

I’d like to turn now to the source of

happiness that is mentioned most often
by people–one’s material living level,
or standard of living. Does more money
make people happier? To judge from
survey responses, most people certainly
think so, although there is a limit. When
asked how much more money they
would need to be completely happy, peo-
ple typically name a ½gure greater than
their current income by about 20 per-
cent. Indeed, if happiness and income
are compared at any point in time, those
with more income are, on average, hap-
pier than those with less.

But what happens to happiness as in-
come goes up over the life cycle–does

happiness go up too? The answer is
no; on average there is no change. Con-
sider, for example, Americans born in
the 1940s. Between the years 1972 and
2000, as their average age increased
from about twenty-six to ½fty-four
years, their average income per person
–adjusted for the change in the price
of goods and services–more than dou-
bled, increasing by 116 percent. Yet their
reported happiness in the year 2000 was
no different from that of twenty-eight
years earlier. They had a lot more money
and a considerably higher standard of
living at the later date, but these did not
make them feel any happier.

Consider, further, two subgroups of
persons born in the 1940s–those with at
least some college education and those
with only a secondary education or less.
At any given age, the more educated sub-
group is happier than the less educated.
This is consistent with the point-of-time
relation between happiness and income
just mentioned–the more educated
being, on average, more affluent and
happier.

But what happens over the life course
to the two educational groups? As one
might expect, the income of the more
educated increases more than that of the
less educated. If happiness were moving
in accordance with the income of each
group, then the happiness of both
groups should increase, with that of the
more educated increasing more, and the
difference between the two groups wid-
ening. In fact, over the life course happi-
ness remains constant for both educa-
tional groups, and the happiness differ-
ential is unchanged. Although those for-
tunate enough to start out with higher
income and education remain, on aver-
age, happier than those of lower socio-
economic status, there is no evidence
that happiness increases with income
growth for either group.

The
economics
of happiness

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
3
2
2
6
1
8
2
8
7
8
8
0
0
1
1
5
2
6
0
4
3
2
3
0
4
9
3
6
1
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Dædalus Spring 2004

Richard A.
Easterlin
on
happiness

These results–both point-of-time and

life cycle–hold as well for people born
in the 1950s, 1930s, and 1920s. Although
the point-of-time result seemingly con-
½rms the economists’ assumption that
more money makes you happier, the life
cycle result contradicts it.

Why this paradoxical pattern? A sim-

ple thought experiment brings out the
basic reason. Imagine your income in-
creases substantially while everyone
else’s stays the same–would you feel
better off? The answer most people
give is yes. But now, let’s turn the exam-
ple around. Think about a situation in
which your real income stays the same
but everyone else’s increases substan-
tially–then how would you feel? Most
people say that they would feel less well-
off, even though their real level of living
hasn’t, in fact, changed at all.

This thought experiment demon-
strates that, as far as material things are
concerned, one’s satisfaction with life
depends not simply on one’s objective
condition, but also on a comparison be-
tween one’s objective condition and a
subjective (or ‘internalized’) living level
norm–and this norm is signi½cantly af-
fected by the average living level of the
people around us. Over time, as every-
one’s income increases, so too do the in-
ternal norms by which we are making
our judgments about happiness. The in-
crease in internal norms is greater for
those with higher income because over
the life cycle we increasingly compare
ourselves against those with whom we
come in closest contact, and contacts are
more and more limited to those of simi-
lar income. The increase over time in
one’s internal living level norm, howev-
er, undercuts the effect of increased ac-
tual income on well-being.

The subversive effect of rising internal

norms also explains why most people
think that over the life course more

money will make them happier when,
in fact, it doesn’t. What actually hap-
pens, of course, is that when their own
income increases, so too does everyone
else’s. This means the internal living
level norms used to evaluate happiness
also increase, offsetting the effect of
growth in their actual income, and so
their happiness stays the same. Here, at
last, we seemingly have a validation of
the psychologists’ model: in the material
goods domain there does appear to be
complete hedonic adaptation.

The survey evidence indicates that over

the life cycle, family and health circum-
stances typically have lasting effects on
happiness, but that more money does
not. What do these empirical results im-
ply for the possibility of increasing one’s
happiness?

Each of us has only a ½xed amount of
time available for family life, health ac-
tivities, and work. Do we distribute our
time in the way that maximizes our hap-
piness? The answer, I believe, is no, for a
reason that has already been suggested:
we decide how to use our time based on
the false belief that more money will
make us happier. Because of this ‘mon-
ey illusion,’ we allocate an excessive
amount of time to monetary goals, and
shortchange nonpecuniary ends such as
family life and health.

As evidence of the perverse effect of
the money illusion, let me cite a survey
reported by sociologist Norval Glenn. In
this survey Americans were asked about
the likelihood of their taking a more re-
warding job that would take away family
time, because it would require both
more hours at the of½ce and more time
on the road. Choosing from four re-
sponse options, not one of the twelve
hundred respondents said it was “very
unlikely” that he or she would take the
job, and only about one in three said it

Dædalus Spring 2004

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
3
2
2
6
1
8
2
8
7
8
8
0
0
1
1
5
2
6
0
4
3
2
3
0
4
9
3
6
1
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

was “somewhat unlikely.” The large ma-
jority of respondents said it was either
“very likely” or “somewhat likely”–
each of these categories accounting for
about one-third of the respondents.
Most Americans, it seems, would readily
sacri½ce family life for what they think
will be greater rewards from their work-
ing life–not knowing that these pro-
spective rewards are likely illusory.

Some may feel that I have given too lit-
tle attention here to the genetic and per-
sonality determinants of happiness. This
is so, but there is a reason. There is noth-
ing one can do, at least at present, about
one’s genes–and very little that most of
us can do about our personalities (ex-
cept, perhaps, to consult a psychologist).
But all of us have the potential for man-
aging our lives more ef½ciently to
achieve greater happiness.

In my discussion of life events, I have

focused on the three–money, family,
and health–that people cite most often
as important for their happiness. I have
tried to summarize here what social sur-
veys have to say about these principal
sources of personal happiness. Could we
make our lives happier? The tentative
answer, based on the evidence at hand,
I suggest, is this. Most people could in-
crease their happiness by devoting less
time to making money, and more time
to nonpecuniary goals such as family life
and health.

The
economics
of happiness

D
o
w
n
o
a
d
e
d

f
r
o
m
h

t
t

:
/
/

d
i
r
e
c
t
.

i
t
.

e
d
u
d
a
e
d
a
r
t
i
c
e
–
p
d

f
/

1
3
3
2
2
6
1
8
2
8
7
8
8
0
0
1
1
5
2
6
0
4
3
2
3
0
4
9
3
6
1
p
d

b
y
g
u
e
s
t

o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3

Dædalus Spring 2004

33
Download pdf