High income improves evaluation of life but not emotional well-being

Author(s): Daniel Kahneman and Angus Deaton

Source: Proceedings of the National Academy of Sciences of the United States of
America , September 21, 2010, Vol. 107, No. 38 (September 21, 2010), pp. 16489-16493
Published by: National Academy of Sciences

Stable URL: https://www.jstor.org/stable/20779694

REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/20779694?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

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 [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

National Academy of Sciences is collaborating with JSTOR to digitize, preserve and extend
access to Proceedings of the National Academy of Sciences of the United States of America

This content downloaded from

13.234.121.200 on Sat, 26 Jun 2021 03:17:33 UTC
All use subject to https://about.jstor.org/terms
 High income improves evaluation of life but not
emotional well-being
Daniel Kahneman1 and Angus Deaton
Center for Health and Well-being, Princeton University, Princeton, NJ 08544

Contributed by Daniel Kahneman, August 4, 2010 (sent for review July 4, 2010)

Recent research has begun to distinguish two aspects of subjective best possible life for you." We find that emotional well-being and
welt-being. Emotional well-being refers to the emotional quality of an life evaluation have different correlates in the circumstances of
individual's everyday experience?the frequency and intensity of ex people's lives. In particular, we observe striking differences in the
periences of joy, stress, sadness, anger, and affection that make one's relationship of these aspects of well-being to income. (For related
life pleasant or unpleasant. Life evaluation refers to the thoughts that observations in the Gallup World Poll, see ref. 9.)
people have about their life when they think about it. We raise the Confusion abounds in discussions of our question. For an ex
question of whether money buys happiness, separately for these two ample, consider the statement that "a lasting marriage.. .is esti
aspects of well-being. We report an analysis of more than 450,000 re mated to be worth $100,000 a year" (10). This correct statement of
sponses to the Gallup-Healthways Well-Being Index, a daily survey of a research finding is likely to be misunderstood, because many
1,000 US residents conducted by the Gallup Organization. We find readers will interpret it by imagining the pleasure of a change of
that emotional well-being (measured by questions about emotional this magnitude in their income. The pleasure of a raise is likely to
experiences yesterday) and life evaluation (measured by Cantril's Self be transient, however, due to a phenomenon known as adapta
Anchoring Scale) have different correlates. Income and education are tion. Because of adaptation, the difference in well-being between
more closely related to life evaluation, but health, care giving, loneli two random individuals whose income differs by $100,000 is far
ness, and smoking are relatively stronger predictors of daily emotions. less impressive than the joy and misery that these individuals
When plotted against log income, life evaluation rises steadily. Emo would immediately experience were they to trade places. Because
tional well-being also rises with log income, but there is no further the observed effects of long-established income differences are
progress beyond an annual income of -$75,000. Low income exacer much smaller than intuitively expected, they are sometimes de
bates the emotional pain associated with such misfortunes as divorce, scribed as inconsequential, but this too is misleading. When en
ill health, and being alone. We conclude that high income buys life sat tered in multiple regression model to predict well-being along
isfaction but not happiness, and that low income is associated both with other aspects of life circumstances (marital status, age, ed
with low life evaluation and low emotional well-being. ucation), the effects of household income are almost invariably
both statistically significant and quantitatively important. We re
life evaluation | emotional experience | household income | satiation | port that household income matters for both emotional well-being
happiness and life evaluation, and that there are circumstances under which
it matters for the latter when it does not matter for the former.
Some of the confusion regarding the effects of income on well
The question of whether
quently in discussions of subjective"money
well-being in buys happiness" comes up fre
both scholarly being can be traced to incorrect analysis. Psychologists and soci
debates and casual conversation. The topic has been addressed in ologists often plot measures of subjective well-being against income
a vast and inconclusive research literature (for a selection of recent in dollars, but a strong argument can be made for the logarithm of
reviews, see refs. 1-4). No single article can settle this complex ques income as the preferred scale. The logarithmic transformation re
tion definitively, but data recently collected by the Gallup Organi presents a basic fact of perception known as Weber's Law, which
zation in the Gallup-Healthways Well-Being Index (GHWBI) pro applies generally to quantitative dimensions of perception and
vide a rich source of observations, as well as an unusually detailed judgment (e.g., the intensity of sounds and lights). The rule is that
measurement of well-being. We analyze the responses of more than the effective stimulus for the detection and evaluation of changes or
450,000 US residents surveyed in 2008 and 2009 to several questions differences in such dimensions is the percentage change, not its ab
about their subjective well-being. The results suggest a rather com solute amount. In the context of income, a $100 raise does not have
plex answer to our opening question. the same significance for a financial services executive as for an in
A discussion of subjective well-being must recognize a dis dividual earning the minimum wage, but a doubling of their re
tinction between two concepts that are often confounded (5-8), spective incomes might have a similar impact on both. The loga
Emotional well-being (sometimes called hedonic well-being or rithmic transformation reveals an important regularity of judgment
experienced happiness) refers to the emotional quality of an that risks being masked when a dollar scale is used.
individual's everyday experience?the frequency and intensity of Plots of subjective well-being against income in dollars in
experiences of joy, fascination, anxiety, sadness, anger, and af variably yield a strongly concave function. Although concavity is
fection that make one's life pleasant or unpleasant. Life evalu entailed by the psychophysics of quantitative dimensions, it often
ation refers to a person's thoughts about his or her life. Surveys has been cited as evidence that people derive little or no psy
of subjective well-being have traditionally emphasized life eval chological benefit from income beyond some threshold. Al
uation. The most commonly asked question in these surveys is though this conclusion has been widely accepted in discussions of
the life satisfaction question: "How satisfied are you with your the relationship between life evaluation and gross domestic pro
life as a whole these days?" The GHWBI survey is unusual in its duct (GDP) across nations (11-14), it is false, at least for this
attempt to distinguish and capture both aspects of subjective
well-being. Emotional well-being is assessed by questions about
the presence of various emotions in the experience of yesterday Author contributions: D.K. and A.D. designed research; performed research; analyzed
data; and wrote the paper.
(e.g., enjoyment, happiness, anger, sadness, stress, worry). Life
The authors declare no conflict of interest.
evaluation is measured using Cantril's Self-Anchoring Scale,
which has the respondent rate his or her current life on a ladder Freely available online through the PNAS open access option.

scale in which 0 is "the worst possible life for you" and 10 is "the 1To whom correspondence should be addressed. E-mail: [email protected].

www.pnas.org/cgi/doi/10.1073/pnas.1011492107 PNAS j September 21,2010 | vol.107 | no. 38 | 16489-16493

This content downloaded from

13.234.121.200 on Sat, 26 Jun 2021 03:17:33 UTC
All use subject to https://about.jstor.org/terms
aspect of subjective well-being. In accordance with Weber's Law, population ranks high on the ladder (ninth after the Scandinavian
average national life evaluation is linear when appropriately plotted countries, Canada, The Netherlands, Switzerland, and New Zea
against log GDP (15); a doubling of income provides similar incre land), and also does well in terms of happiness (fifth), smiling
ments of life evaluation for countries rich and poor. As this example (33rd), and enjoyment (10th), but much less well on worry (89th
illustrates, the statement that "money does not buy happiness" may from best), sadness (69th from best), and anger (75th). Americans
be inferred from a careless reading of a plot of life evaluation against report very high levels of stress (fifth among 151 countries).
raw income?an error avoided by using the logarithm of income. In Table 1 presents regressions of the four well-being measures on
the present study, we confirm the contribution of higher income to a set of demographic variables, which provide context for inter
improving individuals' life evaluation, even among those who are
preting these measures. All of the predictors are dichotomous.
already well off. However, we also find that the effects of income on
The first row of the table shows the regression coefficient for an
the emotional dimension of well-being satiate fully at an annual
indicator of high income, defined as reporting a monthly income
income of ~$75,000, a result that is, of course, independent of
of at least $4,000, which corresponds to the top 58% of the pop
whether dollars or log dollars are used as a measure of income.
ulation. These coefficients cannot be compared across the row,
The aims of our analysis of the GHWBI were to examine pos because the outcom?s have different scales. The entries in other
sible differences between the correlates of emotional well-being
and of life evaluation, focusing in particular on the relationship rows are ratios normalized by the coefficient on the high-income
between these measures and household income. indicator, thus representing the estimated effect relative to the
effect of increasing income by approximately 4-fold. The sign of
Results each ratio is positive if its regression coefficient has the same sign
Some observations were deleted to eliminate likely errors in the as the coefficient for income (positive for positive affect, negative
reports of income. The GHWBI asks individuals to report their for blue affect, etc.). A coefficient >1 indicates an effect larger
monthly family income in 11 categories. The three lowest cate
gories?-0, <$60, and $60-$499?cannot be treated as serious
estimates of household income. We deleted these three catego Table 1. Life evaluation, emotional well-being, income, and the
income-normalized effects of other correlates
ries (a total of 14,425 observations out of 709,183), as well as
those respondents for whom income is missing (172,677 obser Positive affect Blue affect Stress Ladder
vations). We then regressed log income on indicators for the
Regression coefficient
congressional district in which the respondent lived, educational
categories, sex, age, age squared, race categories, marital status High income 0.03 -0.06 -0.03 0.64
Ratio of coefficient to log income coefficient
categories, and height. Thus, we predict the log of each indi
vidual's income by the mean of log incomes in his or her con High income 1.00 1.00 1.00 1.00
Insured 0.40 0.92 1.19 0.59
gressional district, modified by personal characteristics. This re
Old 0.79 0.93 6.28 0.50
gression explains 37% of the variance, with a root mean square
Graduate 0.03 0.01 -1.93 0.48
error (RMSE) of 0.67852. To eliminate outliers and implausible
income reports, we dropped observations in which the absolute Religious 1.16 -0.02 1.21 0.35
Female 0.16 -0.60 -1.89 0.29
value of the difference between log income and its prediction
Married 0.66 0.45 0.66 0.32
exceeded 2.5 times the RMSE. This trimming lost 14,510 obser
Weekend 1.13 0.72 4.83 0.01
vations out of 450,417, or 3.22%. In all, we lost 28.4% of the
Children 0.08 -0.37 -2.47 -0.11
original sample. In comparison, the US Census Bureau imputed
income for 27.5% of households in the 2008 wave of the Ameri Caregiver -0.49 -1.02 -2.99 -0.25
Obese -0.38 -0.14 -0.42 -0.31
can Community Survey (ACS). As a check that our exclusions do
Divorced -0.38 -0.27 -0.88 -0.32
not systematically bias income estimates compared with Census
Health condition -1.36 -1.22 -3.15 -0.48
Bureau procedures, we compared the mean of the logarithm of
Headache -4.45 -3.41 -9.82 -0.78
income in each congressional district from the GHWBI with the
Alone -7.13 -2.10 -3.73 -0.75
logarithm of median income from the ACS. If income is approxi
Smoker -1.01 -0.84 -2.85 -0.70
mately lognormal, then these should be close. The correlation was
0.961, with the GHWBI estimates about 6% lower, possibly attrib All correlates are dichotomous. The first row reports the coefficient of an
utable to the fact that the GHWBI data cover both 2008 and 2009. indicator for high income in regressions of the ladder, positive affect, blue
We defined positive affect by the average of three dichotomous affect, and stress on all correlates. Note that the four outcomes are on different
items (reports of happiness, enjoyment, and frequent smiling and scales. High income is the 58% of the sample whose monthly income is at least
laughter) and what we refer to as "blue affect"?the average of $4,000/mo. The subsequent rows give the regression coefficients on the other
correlates divided by the regression coefficient on the high-income indicator,
worry and sadness. Reports of stress (also dichotomous) were an
and thus show the estimated effect relative to the effect of increasing income
alyzed separately (as was anger, for which the results were similar
by approximately 4-fold. Income has a benef icial effect on all outcomes, so the
but not shown) and life evaluation was measured using the Cantril ratios in rows other than the first are positive when the correlate is associated
ladder. The correlations between the emotional well-being meas with a good effect on the outcome and negative otherwise. "Insured" indi
ures and the ladder values had the expected sign but were modest cates that the respondent has health insurance. "Old" is age 60 y or above.
in size (all <0.31). Positive affect, blue affect, and stress also were "Graduate" indicates a college degree. "Religious" indicates that the respon
weakly correlated (positive and blue affect correlated -0.38, and dent reports that religion is an important part of his or her daily life. "Week
-0.28, and 0.52 with stress.) The results shown here are similar when end" indicates that the day reported on was a Saturday or a Sunday; this is
the constituents of positive and blue affect are analyzed separately. the previous day for stress and for positive and blue affect, and the day of the
interview for the ladder. "Children" is 1 if there are children living in the
As in other studies of well-being, we found that most people
household, and "caregiver" is 1 if the respondent currently helps care for an
were quite happy and satisfied with their lives. About 85% of
elderly or disabled family member, relative, or friend. "Obese" is 1 if body mass
respondents experienced much positive affect (the average of index (based on self-reported height and weight) is >30. "Health condition" is
smiling, enjoyment, and happiness) each day. Blue affect (sadness 1 if the respondent reports ever having been diagnosed by a doctor or nurse
and worry) was reported by 24%, and stress was reported by 39%. with one or more of the following: high blood pressure, high cholesterol, di
The average of the Cantril ladder score was 6.76. Compared with abetes, myocardial infarction, asthma, cancer, or other chronic condition.
about 150 other countries for which we have corresponding data "Alone" is 1 if the respondent reports zero social time with friends or family
from the Gallup World Poll, these results indicate that the US yesterday, including telephone and e-mail contact.

16490 I www.pnas.org/cgi/doi/10.1073/pnas.1011492107 Kahneman and Deaton

This content downloaded from

13.234.121.200 on Sat, 26 Jun 2021 03:17:33 UTC
All use subject to https://about.jstor.org/terms
 than that of the income dichotomy. Because higher incomes are Table 2. Tests for income satiation of life evaluation and
always associated with better outcomes, positive ratios indicate emotional well-being
that the predictor is associated with better outcomes, and negative Positive affect Blue affect Stress Ladder
ratios indicate the opposite.
With few exceptions, the various predictors have the same sign Top vs. second 0.0035
0.0013 0.0055 0.2264
for all four well-being measures, but their relative sizes vary con t value (1.9) (0.6) (1.5) (19.4)
Second vs. third 0.0082 -0.0131 0.0016 0.2268
siderably. As might be expected, weekends are associated with
improved affect, especially with reduced stress. Physical illness, t value (4.4) (5.7) (0.4) (19.7)
Observations
headaches, spending a day alone, and caring for an adult all have
relatively larger adverse effects on emotional well-being than on
Top group 72,744 73,104 73,109 73,068
life evaluation. Headaches and being alone, like emotional well
Second group 40,136 40,291 40,301 40,283
being, are measured for yesterday, which may enhance their im
Third group 88,887 89,278 89,290 89,245
portance in the regressions. At the other extreme, being a college The coefficients reported are the differences in mean outcomes between the
graduate is associated with high life evaluation but has only a small two indicated income categories. The top category is >$10,000/mo, the second
association with positive and blue affect and a (perhaps) coun category is $7,500-$9,999/mo, and the third category $5,000-$7,499/mo. SEs are
terintuitive relation with stress; all other factors being equal, col corrected for spatial clustering within zip codes.

lege graduates report more stress than nongraduates. The Gallup

World Poll found high levels of stress in high-GDP countries (16).
Religion has a substantial influence on improving positive af
a log scale. (The midpoints of each income range, used only in
the figure, are imputed assuming that the underlying distribution
fect and reducing reports of stress, but no effect on reducing
of income is lognormal; the figure shows vertical lines for the top
sadness or worry. Females report slightly higher positive affect
and life evaluation, but also more blue affect and much more three interval limits.) Stress is the average of a yes/no response to
the question: "Did you experience a lot of stress yesterday?"
stress. The presence of children at home is associated with sig
Thus, Fig. 1 shows the percentage of the population in each
nificant increases in stress, sadness, and worry (6). As reported
income group who did not report experiencing this emotion on
recently, older people enjoy greater emotional well-being, most
the previous day. Not blue is 1 minus the average of the per
notably a pronounced reduction in the experience of stress and
centage reporting sadness and worry. The right-hand axis shows
anger (17). Smoking is an impressively strong predictor of low the average score on the ladder, with values ranging from 0 to 11.
well-being?especially its emotional dimensions?even when Fig. 1 shows that for all measures of experienced well-being,
income and education are controlled for. A propensity to smoke individuals in the lower- income groups do worse on average than
is in part genetically determined (18) and is a known indicator of those above them, but that those in the top two groups do not
a tense personality (19, 20). differ. For the two top categories to be equal, the entire range of
Fig. 1 and Table 2 characterize the relationship between the the second category must lie above the satiation point. This ob
dimensions of subjective well-being and household income. Fig. servation implies that emotional well-being satiates somewhere in
1 presents averages over eight income groups for the three the third category of income from the top. We infer that beyond
aspects of emotional well-being and for the Cantril ladder about $75,000/y, there is no improvement whatever in any of the
measure of life evaluation. Here blue affect and stress are con
three measures of emotional well-being. In contrast, the figure
verted to their complements, not blue and stress-free, so that shows a fairly steady rise in life evaluation with log income over
higher values in the figure always refer to better psychological the entire range; the effects of income on individuals' life evalu
outcomes. Income is converted to an annual basis and plotted on ations show no satiation, at least to an amount well over $120,000.
Table 2 reports a formal test of satiation for the four measures,
showing how the second-to-top income group (annual income
$90,000-$120,000) differs from the group immediately belo\f it
($60,000-$90,000) and from the group immediately above it
(> $120,000). Positive affect, blue affect, and Cantril ladder score
Positive affect
are all significantly improved in the first comparison with the ex
ception of stress, which appears to satiate at a lower income level,
roughly $60,000. In comparisons of the top two categories, only
the ladder score shows a significant improvement with higher
Not blue y
/
in ?
income. The small t values are remarkable in these very large
CD ?
samples. We conclude that lack of money brings both emotional
/ Ladder misery and low life evaluation; similar results were found for
anger. Beyond ~$75,000 in the contemporary United States, how
ever, higher income is neither the road to experienced happiness
Stress free
nor the road to the relief of unhappiness or stress, although higher
income continues to improve individuals' life evaluations.
Below $75,000, many factors become gradually worse, at least on
average. For example, the emotional pain associated with ill health
10,000 20,000 40,000 80,000 160,000 depends on income; for those reporting a monthly income of at
Annual income least $3,000 (about two-thirds of households), the fractions re
porting blue affect with and without headaches are 38% and 19%,
Fig. 1. Positive affect, blue affect, stress, and life evaluation in relation to
respectively, a difference of 19 percentage points. The correspond
household income. Positive affect is the average of the fractions of the
population reporting happiness, smiling, and enjoyment. "Not blue" is 1 ing values for those with a monthly income of <$1,000 (about 10% of
minus the average of the fractions of the population reporting worry and households) are 70% and 38%, a difference of 32%. Table 3 shows
sadness. "Stress free" is the fraction of the population who did not report that the pain of some of life's misfortunes, including asthma, divorce,
stress for the previous day. These three hedonic measures are marked on the and being alone, is significantly exacerbated by poverty; even the
left-hand scale. The ladder is the average reported number on a scale of benefits of the weekend are less for the poor. Similar results apply to
0-10, marked on the right-hand scale. stress and positive affect.

Kahneman and Deaton PNAS | September 21, 2010 | vol.107 | no. 38 | 16491

This content downloaded from

13.234.121.200 on Sat, 26 Jun 2021 03:17:33 UTC
All use subject to https://about.jstor.org/terms
 Table 3. Poverty exacerbates the effect of adverse that we found in the present study is unusually sharp. We speculate
circumstances: Average percentage of people reporting a lot of that the Cantril ladder of life is a purer measure of life evaluation
sadness and worry yesterday, by income group and condition than the life satisfaction question, which has an emotional aspect,
Income <$1,000/mo Income >$3,000/mo and that the reports of the emotions of yesterday provide a purer
measure of emotional well-being than the standard happiness
No Yes Difference No Yes Difference question. If both aspects of subjective well-being are considered
Weekend 46.6 44.5 -2.1 22.3 17.1 -5.2 important, then the separation of the measures is an advantage.
Divorced 44.3 50.5 6.2 20.5 24.4 3.9 The relevance of subjective well-being as a guide to policy is
Alone 44.0 58.9 14.9 20.5 31.5 11.0 a contentious issue, on which we do not take a position. If
Headache 38.0 69.5 31.6 18.9 38.4 19.5 measures of well-being are to be used to assess human welfare and
Asthma 33.1 40.8 7.8 18.0 21.6 3.6 to guide policy, the present findings raise the question of whether
life evaluation or emotional well-being is better suited to these
aims. Theincome
Approximately 10% of US households have a monthly Cantril ladder is a serious contender for the best tool for
<$ 1,000,
and around two-thirds have a monthly income of measuring the The
>$3,000. degreereported
to which individuals view themselves as
figures are unadjusted for covariates. For the asthma variable,
achieving to control
their goals, both material and other. But emotional
for persons with multiple conditions, who are more common at low incomes,
well-being also is clearly important for individuals and for policy,
the comparison is between those with asthma and no other health condi
and here there are choices as well. Not everyone will agree that
tions and those reporting no health conditions at all. The difference in the
third column is statistically different from the difference in the sixth column
enhancing the happiness experienced by those who are already
quite
in each case; the t values range from 23.9 (headache) tohappy is a legitimate policy objective. The policy goal of
3.1 (asthma.)
reducing suffering is likely to raise fewer objections, and measures
of emotional pain may be useful for that purpose. This topic
Discussion merits serious debate.

The data for positive and blue affect provide an unexpectedly sharp Materials and Methods
answer to our original question. More money does not necessarily The survey involved a telephone interview using a dual-frame random-digit
buy more happiness, but less money is associated with emotional dial methodology that included cell phone numbers from all 50 US states.
pain. Perhaps $75,000 is a threshold beyond which further increases Interviews were conducted between 9:00 AM and 10:00 PM (local time), with
in income no longer improve individuals' ability to do what matters most done in the evening. Up to five callbacks were made in the case of no
most to their emotional well-being, such as spending time with answer. Spanish language interviews were conducted when appropriate.
people they like, avoiding pain and disease, and enjoying leisure. Approximately 1,000 interviews were completed daily from January 2 through
December 30, 2009.
According to the ACS, mean (median) US household income was
$71,500 ($52,000) in 2008, and about a third of households were The questionnaire covered many topics of interest to the Gallup Organi
zation and Healthways Corporation, including basic demographic information,
above the $75,000 threshold. It also is likely that when income rises
participants' opinions about the current economic climate and their personal
beyond this value, the increased ability to purchase positive experi financial situation, information about past diseases, and other topics.
ences is balanced, on average, by some negative effects. A recent Life evaluation was assessed using Cantril's Self-Anchoring Scale (the
psychological study using pinning methods provided suggestive evi ladder), worded as follows: "Please imagine a ladder with steps numbered
dence of a possible association between high income and a reduced from 0 at the bottom to 10 at the top. The top of the ladder represents the
ability to savor small pleasures (21). best possible life for you, and the bottom of the ladder represents the worst
When interpreting our findings, it is essential to distinguish changes possible life for you. On which step of the ladder would you say you per
from differences. Our data speak only to differences; they do not sonally feel you stand at this time?" (15). Questions about emotional well
being had yes/no response options and were worded as follows: "Did you
imply that people will not be happy with a raise from $100,000 to
experience the following feelings during a lot of the day yesterday? How
$150,000, or that they will be indifferent to an equivalent drop in in
about _?" Each of several emotions (e.g., enjoyment, stress) was
come. Changes of income in the high range certainly have emotional reported separately. The positive affect score was the average of the reports
consequences. What the data suggest is that above a certain level of of enjoyment and happiness and of a dichotomous question about the
stable income, individuals' emotional well-being is constrained by frequency of smiling: "Did you smile or laugh a lot yesterday?" The blue
other factors in their temperament and life circumstances. affect score was the average of worry and sadness.
We observe a qualitative difference between our measures of To broaden coverage and representativeness, cell phones were part of the
emotional well-being and of life evaluation?the former satiates sampling design. Relative to land lines, the response rate for cell phones was
with high income, whereas the latter does not. This observation typically lower. Of all calls that resulted in contacts with an eligible candidate,
31% of the candidates agreed to be interviewed; of these, 90% completed
underscores the importance of the distinction between the judg
the entire interview. Despite the sampling limitations, available evidence
ments individuals make when they think about their life and the
suggests that the estimates of population parameters were not compro
feelings that they experience as they live it. As might be expected, the mised; for example, the survey predicted recent election results within an
former is sensitive to socioeconomic status, whereas the latter is acceptable margin of error.
sensitive to circumstances that evoke positive and negative emo
tions, such as spending time with others and caring for a sick relative. ACKNOWLEDGMENTS. We thank Carol Graham, Richard Nisbett, Norbert
Several authors have commented on a related difference between Schwarz, and Arthur Stone for their comments. Special thanks to James
two questions that are often used in surveys of subjective well-being: Harter, one of the authors of the Gallup-Healthways Well-Being Index, for
his contribution to the research and for his comments on this report. We
"How satisfied are you with your life?" and "How happy are you these thank the Gallup Organization and Healthways Corporation for access to the
days?" (8, 22, 23). The common conclusion is that income is more survey results. This work was supported by the Gallup Organization and by
strongly related to satisfaction than to happiness, but the difference National Institute on Aging Grant AG024928-06.

1. Diener E, Biswas-Diener R (2002) Will money increase subjective well-being? Soc Indie 4. Clark AE, Kristensen N, Westergaard-Nielsen (2009) Economic satisfaction and
Res 57:119-169. income rank in small neighbourhoods. J Eur Econ Assoc 7:519-527.
2. Headey , Muffels R, Wooden M (2008) Money does not buy happiness: Or does it? A 5. Diener E (1984) Subjective well-being. Psycho! Bull 95:542-575.
reassessment based on the combined effects of wealth, income and consumption. Soc 6. Kahneman D, Krueger AB, Schkade DA, Schwarz , Stone AA (2004) A survey method for
Indie Res 87:65-82. characterizing daily life experience: The day reconstruction method. Science 306:1776-1780.
3. Clark AE, Frijters P, Shields M (2008) Relative income, happiness and utility: An 7. Kahneman D, Riis J (2005) The Science Of Well-Being, eds Huppert FA, Baylis N,
explanation for the Easterlin paradox and other puzzles. J Econ Lit 46:95-144. Keverne (Oxford Univ. Press, Oxford), pp 285-304.

16492 I wvvw.pnas.org/cgi/doi/10.1073/pnas.1011492107 Kahneman and Deaton

This content downloaded from

13.234.121.200 on Sat, 26 Jun 2021 03:17:33 UTC
All use subject to https://about.jstor.org/terms
8. Graham C (2010) Happiness Around the World: The Paradox of Happy Peasants and 16. Ng W, Diener E, Arora R, Harter J (2009) Affluence, feelings of stress, and well-being.
Miserable Millionaires (Oxford University Press, New York). Soc Indie Res 94:257-271.
9. Diener E, Ng W, Harter J, Arora R (2010) Wealth and happiness across the world: 17. Stone AA, Schwartz JE, Broderick JE, Deaton A (2010) A snapshot of the age distribution of
Material prosperity predicts life evaluation, whereas psychosocial prosperity predicts psychological well-being in the United States. Proc Nati Acad Sci USA 107:9985-9990.
positive feeling. J Pers Soc Psychol 99:52-61. 18. Munaf? MR, Johnstone EC (2008) Genes and cigarette smoking. Addiction 103:893-904.
10. Blanchfl?wer DG, Oswald AJ (2004) Well-being over time in Britain and the USA. 19. Munaf? MR, Zetteler Jl, Clark TG (2007) Personality and smoking status: A meta
J Public Econ 88:1359-1386.
analysis. Nicotine Tob Res 9:405-413.
11. Diener E, Suh EM (1999) Well-Being: The Foundations of Hedonic Psychology, eds 20. Terracciano A, Costa PT, Jr. (2004) Smoking and the five-factor model of personality.
Kahneman D, Diener E, Schwarz (Russell Sage Foundation, New York), pp Addiction 99:472-481.
434-452.
21. Quoidbach J, Dunn EW, Petrides KV, Mikolajczak M (2010) Money giveth, money
12. Layard R (2005) Happiness: Lessons from a New Science (Penguin, New York).
taketh away: The dual effect of wealth on happiness. Psychol Sci 21:759-763.
13. Inglehart R, Klingemann HD (2000) Culture and Subjective Weil-Being, eds Diener E,
22. Howell RT, Howell G (2008) The relation of economic status to subjective well-being
Suh EM (MIT Press, Cambridge, MA), pp 165-184.
14. Veenhoven R (1991) Is happiness relative? Soc Indie Res 24:1-34. in developing countries: A meta-analysis. Psychol Bull 134:536-560.
15. Deaton A (2008) Income, health, and well-being around the world: Evidence from the 23. Veenhoven R, Hagerty M (2006) Rising happiness in nations, 1946-2004: A reply to
Gallup World Poll. J Econ Perspect 22:53-72. Easterlin. Soc Indie Res 79:421-436.

o
3

Kahneman and Deaton PNAS | September 21, 2010 | vol. 107 | no. 38 | 16493

This content downloaded from

13.234.121.200 on Sat, 26 Jun 2021 03:17:33 UTC
All use subject to https://about.jstor.org/terms