World Happiness Report 2016: rapport sur les pays où il fait bon vivre
42 Pages
Downloading requires you to have access to the YouScribe library
Learn all about the services we offer

World Happiness Report 2016: rapport sur les pays où il fait bon vivre

Downloading requires you to have access to the YouScribe library
Learn all about the services we offer
42 Pages


Dans l’édition 2016 du World Happiness report, la France se situe seulement à la 32ème place des pays où il fait bon vivre.



Published by
Published 17 March 2016
Reads 11
Language English
Document size 1 MB


Chapter 2
John F. Helliwell, Canadian Institute for Advanced Research and Vancouver School of Economics,
University of British Columbia8
Haifang Huang, Department of Economics, University of Alberta
Shun Wang, KDI School of Public Policy and Management, Korea
The authors are grateful to the Canadian Institute for Advanced Research and the KDI School for research support, and to
the Gallup Organization for data access and assistance. In particular, several members of the Gallup staff helped in the
development of Technical Box 3. The author are also grateful for helpful advice and comments from Ed Diener, Curtis Eaton,
Carrie Exton, Leonard Goff, Carol Graham, Shawn Grover, Richard Layard, Guy Mayraz, Hugh Shiplett and Conal Smith.W ORLD HAPPINES S REPOR T 20 16 | UPD ATE
survey years 2005 through 2011, in order to Introduction
achieve representative samples in each answer
It is now almost four years since the publication category. In this chapter we repeat that analysis
of the frst World Happiness Report (WHR) in using data from the subsequent four years,
2012. Its central purpose was to survey the 2012-2015. This will give us suffciently large
scientifc underpinnings of measuring and samples to compare what we found for
2005understanding subjective well-being. Its main 2011 with what we now see in the data for
content is as relevant today as it was then, and 2012-2015.
remains available for those now coming to the
topic for the frst time. The subsequent World
Our main analysis of the distribution of
happiHappiness Report 2013 and World Happiness
ness among and within nations continues to be
Report 2015, issued at roughly 18 month
interbased on individual life evaluations, roughly
vals, updated and extended this background. To
1,000 per year in each of more than 150
counmake this World Happiness Report 2016 Update
tries, as measured by answers to the Cantril
accessible to those who are coming fresh to the
ladder question: “Please imagine a ladder, with
World Happiness Report series, we repeat enough
steps numbered from 0 at the bottom to 10 at
of the core analysis in this chapter, and its
the top. The top of the ladder represents the best
several on-line appendices, to explain the
meanpossible life for you and the bottom of the ladder
ing of the evidence we are reporting.
represents the worst possible life for you. On
which step of the ladder would you say you
Chapter 2 in World Happiness Report 2015, the personally feel you stand at this time?” We will,
Geography of Wappiness, started with a as usual, present the average life evaluation
global map, and continued with our attempts to scores for each country, in this report based on
explain the levels and changes in average nation- averages from the surveys conducted in 2013,
al life evaluations among countries around the 2014 and 2015.
world. This year we shall still consider the
geographic distribution of life evaluations
This will be followed, as in earlier editions, by
among countries, while extending our analysis
our latest attempts to show how six key variables
to consider in more detail the inequality of
contribute to explaining the full sample of
happiness – how life evaluations are distributed
national annual average scores over the whole
among individuals within countries and
geoperiod 2005-2015. These variables include GDP
graphic regions.
per capita, social support, healthy life
expectancy, social freedom, generosity and absence of
In studying more deeply the distribution of corruption. We shall also show how measures of
happiness within national and regional popula- experienced well-being, especially positive
tions, we are extending the approach adopted in emotions, can add to life circumstances in the
Chapter 2 of the frst World Happiness Report, in support for higher life evaluations.
which Figure 2.1 showed the global distribution
of life evaluations among the 11 response
categoWe shall then turn to consider the distribution of
ries, with the worst possible life as a 0 and the
9life evaluations among individuals in each
counbest possible life as a 10 (the Cantril ladder
try, using data from all 2012-2015 surveys, with
question). The various parts of Figure 2.2 then
the countries ranked according to the equality of
made the same allocation of responses for
life evaluations among their survey respondents,
respondents in nine global regions, weighting
as measured by the standard deviation from the
the responses from different countries according
mean. We shall then show how these national
to each country’s population. In those fgures we
measures of the equality of life evaluations have
combined all the data then available, for the
changed from 2005-2011 to 2012-2015.Our reason for paying more attention to the Measuring and Understanding
distribution of life evaluations is quite simple. If Happiness
it is appropriate to use life evaluations as an
Chapter 2 of the frst World Happiness Report umbrella measure of the quality of life, to
suppleexplained the strides that had been made during ment and consolidate the benefts available from
the preceding 30 years, mainly within psychology, income, health, family and friends, and the
in the development and validation of a variety of broader institutional and social context, then it is
measures of subjective well-being. Progress since equally important to broaden the measurement
then has moved faster, as the number of scientifc of inequalities beyond those for income and
papers on the topic has continued to grow wealth. Whether people are more concerned with
1rapidly, and as the measurement of subjective equality of opportunities or equality of outcomes,
well-being has been taken up by more national the data and analysis should embrace the
availand international statistical agencies, guided by ability of and access to sustainable and livable
technical advice from experts in the feld. cities and communities as much as to income
and wealth. We will make the case that the
distribution of life evaluations provides an By the time of the frst report there was already
over-arching measure of inequality in just the a clear distinction to be made among three main
same way as the average life evaluations provide classes of subjective measures: life evaluations,
an umbrella measure of well-being. positive emotional experiences (positive affect)
and negative emotional experiences (negative
affect); see Technical Box 1. The Organization The structure of the chapter is as follows. We
for Economic Co-operation and Development shall start with a review of how and why we use
(OECD) subsequently released Guidelines on life evaluations as our central measure of
subjec2Measuring Subjective Well-being, which included tive well-being within and among nations. We
both short and longer recommended modules of shall then present data for average levels of life
3subjective well-being questions. The centerpiece evaluations within and among countries and
of the OECD short module was a life evaluation global regions. This will include our latest
question, asking respondents to assess their efforts to explain the differences in national
satisfaction with their current lives on a 0 to 10 average evaluations, across countries and over
scale. This was to be accompanied by two or the years. After that we present the latest data on
three affect questions and a question about the changes between 2005-2007 and 2013-2015 in
extent to which the respondents felt they had average national life evaluations.
a purpose or meaning in their lives. The latter
question, which we treat as an important
supWe shall then turn to consider inequality and
port for subjective well-being, rather than a well-being. We frst provide a country ranking of
4direct measure of it, is of a type that has come
the inequality of life evaluations based on data
to be called “eudaimonic,” in honor of Aristotle, from 2012-2015, followed by a country ranking
who believed that having such a purpose would
based on the size of the changes in inequality
be central to any refective individual’s assess-that have taken place between 2005-2011 and
ment of the quality of his or her own life.
2012-2015. We then attempt to assess the possible 10
consequences for average levels of well-being,
Chapter 2 of World Happiness Report 2015 re-and for what might be done to address well-being
viewed evidence from many countries and inequalities. We conclude with a summary of our
several different surveys about the types of latest evidence and its implications.
information available from different measures
8of subjective well-being. What were the main
messages? First, all three of the commonly used W ORLD HAPPINES S REPOR T 20 16 | UPD ATE
Technical Box 1: Measuring Subjective Well-being
The OECD (2013) Guidelines on Measuring Sub- The second element consists of a short series of
jective Well-being, quotes in its introduction the affect questions and an experimental
eudaimonfollowing defnition and recommendation from ic question (a question about life meaning or
the earlier Commission on the Measurement of purpose). The inclusion of these measures
comEconomic and Social Progress: plements the primary evaluative measure both
because they capture different aspects of
subjec“Subjective well-being encompasses three dif- tive well-being (with a set of drivers)
ferent aspects: cognitive evaluations of one’s and because the difference in the nature of the
life, positive emotions (joy, pride), and nega- measures means that they are affected in
differtive ones (pain, anger, worry). While these as- ent ways by cultural and other sources of
meapects of subjective well-being have different surement error. While it is highly desirable that
determinants, in all cases these determinants these questions are collected along with the
prigo well beyond people’s income and material mary measure as part of the core, these
quesconditions... All these aspects of subjective tions should be considered a lower priority than
6well-being should be measured separately to the primary measure.”
derive a more comprehensive measure of
peo7ple’s quality of life and to allow a better under- Almost all OECD countries now contain a life
standing of its determinants (including peo- evaluation question, usually about life
satisfacple’s objective conditions). National statistical tion, on a 0 to 10 rating scale, in one or more of
agencies should incorporate questions on sub- their surveys. However, it will be many years
bejective well-being in their standard surveys to fore the accumulated efforts of national
statisticapture people’s life evaluations, hedonic expe- cal offces will produce as large a number of
5riences and life priorities.” comparable country surveys as is now available
through the Gallup World Poll (GWP), which
The OECD Guidelines go on to recommend a has been surveying an increasing number of
core module of questions to be used by national countries since 2005, and now includes almost
statistical agencies in their household surveys: all of the world’s population. The GWP contains
one life evaluation as well as a range of positive
“There are two elements to the core measures and negative experiential questions, including
module. several measures of positive and negative affect,
mainly asked with respect to the previous day.
The frst is a primary measure of life evaluation. In this chapter, we make primary use of the life
This represents the absolute minimum re- evaluations, since they are, as we show in Table
quired to measure subjective well-being, and it 2.1, more international in their variation and are
is recommended that all national statistical more readily explained by life circumstances.
agencies include this measure in one of their
annual household surveys.
life evaluations (specifcally Cantril ladder, respondents’ answers to the Cantril ladder
satisfaction with life, and happiness with life in question, with its use of a ladder as a framing
general) tell almost identical stories about the device, were more dependent on their incomes
nature and relative importance of the various than were answers to questions about
satisfacfactors infuencing subjective well-being. For tion with life. The evidence for this came from
example, for several years it was thought (and is comparing modeling using the Cantril ladder in
still sometimes reported in the literature) that the Gallup World Poll (GWP) with modeling based on life satisfaction answers in the World answers? For this important question, no
defniValues Survey (WVS). But this conclusion, based tive answer was available until the European
on comparing two different surveys, unfortu- Social Survey (ESS) asked the same respondents
nately combines survey and method differences “satisfaction with life” and “happy with life”
with the effects of question wording. When it questions, wisely using the same 0 to 10
resubsequently became possible to ask both sponse scales. The answers showed that income
9questions of the same respondents on the and other key variables all have the same effects
same scales, as was the case in the Gallup on the “happy with life” answers as on the
World Poll in 2007, it was shown that the “satisfed with life” answers, so much so that
estimated income effects and almost all other once again more powerful explanations come
structural infuences were identical, and a more from averaging the two answers.
powerful explanation was obtained by using an
10average of the two answers. Another previously common view was that
changes in life evaluations at the individual level
It was also believed at one time that when were largely transitory, returning to their
basequestions included the word “happiness” they line as people rapidly adapt to their
circumstancelicited answers that were less dependent on es. This view has been rejected by four
indepenincome than were answers to life satisfaction dent lines of evidence. First, average life
questions or the Cantril ladder. Evidence for that evaluations differ signifcantly and
systematicalview was based on comparing World Values ly among countries, and these differences are
11Survey happiness and life satisfaction answers, substantially explained by life circumstances.
and by comparing the Cantril ladder with happi- This implies that rapid and complete adaptation
ness yesterday (and other emotions yesterday). to different life circumstances does not take
Both types of comparison showed the effects of place. Second, there is evidence of long-standing
income on the happiness answers to be less trends in the life evaluations of sub-populations
signifcant than on satisfaction with life or the within the same country, further demonstrating
Cantril ladder. Both conclusions were based on that life evaluations can be changed within
13the use of non-comparable data. The frst com- policy-relevant time scales. Third, even though
parison, using WVS data, involved different individual-level partial adaptation to major life
scales and a question about happiness that events is a normal human response, there is
might have combined emotional and evaluative very strong evidence of continuing infuence on
components. The second strand of literature, well-being from major disabilities and
unem14based on GWP data, compared happiness ployment, among other life events. The case of
yesterday, which is an experiential/emotional marriage is still under debate. Some recent
response, with the Cantril ladder, which is results using panel data from the UK have
equally clearly an evaluative measure. In that suggested that people return to baseline levels of
context, the fnding that income has more life satisfaction several years after marriage, a
purchase on life evaluations than on emotions result that has been argued to support the more
15seems to have general applicability, and stands general applicability of set points. However,
12as an established result. subsequent research using the same data has
shown that marriage does indeed have
long-lasting well-being benefts, especially in protecting But what if happiness is used as part of a life
the married from as large a decline in the evaluation? That is, if respondents are asked
middle-age years that in many countries repre-how happy, rather than how satisfed, they are
16sent a low-point in life evaluations. Fourth, and with their life as a whole? Would the use of
especially relevant in the global context, are “happiness” rather than “satisfaction” affect the
studies of migration showing migrants to have infuence of income and other factors on the W ORLD HAPPINES S REPOR T 20 16 | UPD ATE
average levels and distributions of life evalua- people see through the day-to-day and
hour-totions that resemble those of other residents of hour fuctuations, so that the answers they give
their new countries more than of comparable on weekdays and weekends do not differ.
residents in the countries from which they have
17emigrated. This confrms that life evaluations On the other hand, although life evaluations do
do depend on life circumstances, and are not not vary by the day of week, they are much more
destined to return to baseline levels as required responsive than emotional reports to differences
by the set point hypothesis. in life circumstances. This is true whether the
20comparison is among national averages or
21among individuals.
Why Use Life Evaluations for
Furthermore, life evaluations vary more between International Comparisons of
countries than do emotions. Thus almost the Quality of Life?
one-quarter of the global variation in life
evaluaIn each of the three previous World Happiness tions is among countries, compared to
Reports we presented different ranges of data three-quarters among individuals in the same
covering most of the experiences and life evalua- country. This one-quarter share for life
evaluations that were available for a large number of tions is far more than for either positive affect
countries. We were grateful for the breadth of (7 percent) or negative affect (4 percent). This
available information, and used it to deepen our difference is partly due to the role of income,
understanding of the ways in which experiential which plays a stronger role in life evaluations
and evaluative reports are connected. Our than in emotions, and is also very unequally
conclusion is that while experiential and evalua- spread among countries. For example, more
tive measures differ from each other in ways than 40 percent of the global variation among
that help to understand and validate both, life household incomes is among nations rather
22evaluations provide the most informative mea- than among individuals within nations.
sures for international comparisons because
they capture the overall quality of life as a whole. These twin facts – that life evaluations vary
much more than do emotions across countries,
For example, experiential reports about happi- and that these life evaluations are much more
ness yesterday are well explained by events of fully explained by life circumstances than are
the day being asked about, while life evaluations emotional reports– provide for us a suffcient
more closely refect the circumstances of life as a reason for using life evaluations as our central
whole. Most Americans sampled daily in the measure for making international
compari23Gallup-Healthways Well-Being Index Survey feel sons. But there is more. To give a central role
happier on weekends, to an extent that depends to life evaluations does not mean we need to
on the social context on and off the job. The either ignore or downplay the important
inforweekend effect disappears for those employed in mation provided by experiential measures. On
a high trust workplace, who regard their superi- the contrary, we see every reason to keep
experi13or more as a partner than a boss, and maintain ential measures of well-being, as well as
mea18their social life during weekdays. sures of life purpose, as important elements in
our attempts to measure and understand
subjective well-being. This is easy to achieve, at least in By contrast, life evaluations by the same
responprinciple, because our evidence continues to dents in that same survey show no weekend
19 suggest that experienced well-being and a sense effects. This means that when they are
answerof life purpose are both important infuences on ing the evaluative question about life as a whole, life evaluations, above and beyond the critical In Table 2.1 we present our latest modeling of
role of life circumstances. We shall provide national average life evaluations and measures
direct evidence of this, and especially of the of positive and negative affect (emotion) by
importance of positive emotions, in Table 2.1. country and year. For ease of comparison, the
Furthermore, in Chapter 3 of World Happiness Table has the same basic structure as Table 2.1 in
Report 2015 we gave experiential reports a central the World Happiness Report 2015. The major
role in our analysis of variations of subjective difference comes from the inclusion of data for
well-being across genders, age groups, and late 2014 and 2015, which increases by 144 (or
global regions. about 15 percent) the number of country-year
25observations. The resulting changes to the
26estimated equation are very slight. There are We would also like to be able to compare
infour equations in Table 2.1. The frst equation equality measures for life evaluations with those
provides the basis for constructing the sub-bars for emotions, but unfortunately that is not
shown in Figure 2.2. currently possible, since the Gallup World Poll
emotion questions all offer only yes and no
responses. Thus nothing can be said about their The equation explains national average life
distribution beyond the national average shares evaluations in terms of six key variables: GDP
of yes and no answers. For life evaluations, per capita, social support, healthy life
expectanhowever, there are 11 response categories, so we cy, freedom to make life choices, generosity and
27are able to contrast distribution shapes for each freedom from corruption. Taken together,
country and region, and see how these evolve as these six variables explain almost three-quarters
time passes. We start by looking at the popula- of the variation in national annual average
tion-weighted global and regional distributions ladder scores among countries, using data from
of life evaluations, based on how respondents the years 2005 to 2015. The model’s predictive
24rate their lives . power is little changed if the year fxed effects in
the model are removed, falling from 74.1% to
73.6% in terms of the adjusted r-squared. In the rest of this report, Cantril ladder is the
only measure of life evaluations to be used, and
“happiness” and “subjective well-being” are used Figure 2.1: Population-Weighted Distributions of
exchangeably. All the analysis on the levels or Happiness, 2012-2015 (Part 1)
changes of subjective well-being refers only to
life evaluations, specifcally the Cantril ladder.
Mean = 5.353
SD = 2.243
The Distribution of Happiness
around the World .2
The various panels of Figure 2.1 contain bar
charts showing for the world as a whole, and for
14 each of 10 global regions, the distribution of the .1
2012-2015 answers to the Cantril ladder question
.05asking respondents to value their lives today on
a 0 to 10 scale, with the worst possible life as a 0
0 1 2 3 4 5 6 7 8 9 10 and the best possible life as a 10.
Figure 2.1: Population-Weighted Distributions of Happiness, 2012-2015 (Part 2)
.35 .35
Mean = 7.125 Mean = 6.578
SD = 2.016 SD = 2.329.3 .3
.25 .25
.2 .2
.15 .15
.1 .1
.05 .05

0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
Northern America & ANZ Latin America & Caribbean
.35 .35
Mean = 6.575 Mean = 5.554
SD = 1.944 SD = 2.152.3 .3
.25 .25
.2 .2
.15 .15
.1 .1
.05 .05
0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
Western Europe Central and Eastern Europe
.35 .35
Mean = 5.502 Mean = 5.363
SD = 2.073 SD = 1.963.3 .3
.25 .25
.2 .2
.15 .15
.1 .1
.05 .05
0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
Commonwealth of Independent States Southeast Asia
.35 .35
Mean = 5.288 Mean = 4.999
SD = 2.000 SD = 2.452.3 .3
.25 .25
.2 .2
.15 .15
.1 .1
.05 .05

0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
East Asia Middle East & North Africa
.35 .35
Mean = 4.589 Mean = 4.370
SD = 2.087 SD = 2.115.3 .3
.25 .25
.2 .2
.15 .15
.1 .1
.05 .05
0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
South Asia Sub-Saharan AfricaTable 2.1: Regressions to Explain Average Happiness across Countries (Pooled OLS)
Dependent Variable
Independent Variable Cantril Ladder Positive Affect Negative Affect Cantril Ladder
Log GDP per capita 0.338 -0.002 0.011 0.341
(0.059)*** (0.009) (0.008) (0.058)***
Social support 2.334 0.253 -0.238 1.768
(0.429)*** (0.052)*** (0.046)*** (0.417)***
Healthy life expectancy at birth 0.029 0.0002 0.002 0.028
(0.008)*** (0.001) (0.001)* (0.008)***
Freedom to make life choices 1.056 0.328 -0.089 0.315
(0.319)*** (0.039)*** (0.045)** (0.316)
Generosity 0.820 0.171 -0.011 0.429
(0.276)*** (0.032)*** (0.030) (0.277)
Perceptions of corruption -0.579 0.033 0.092 -0.657
(0.282)** (0.030) (0.025)*** (0.271)**
Positive affect 2.297
Negative affect 0.050
Year fxed effects Included Included Included Included
Number of countries 156 156 156 156
Number of observations 1,118 1,115 1,117 1,114
Adjusted R-squared 0.741 0.497 0.226 0.765
Notes: This is a pooled OLS regression for a tattered panel explaining annual national average Cantril ladder
responses from all available surveys from 2005 to 2015. See Technical Box 2 for detailed information about each
of the predictors. Coeffcients are reported with robust standard errors clustered by country in parentheses.
***, **, and * indicate signifcance at the 1, 5 and 10 percent levels respectively.
The second and third columns of Table 2.1 use Bearing in mind that positive and negative affect
the same six variables to estimate equations for are measured on a 0 to 1 scale, while life
evaluanational averages of positive and negative affect, tions are on a 0 to 10 scale, social support can be
where both are based on averages for answers seen to have a similar proportionate effect on
about yesterday’s emotional experiences. In positive and negative emotions as on life
evaluageneral, the emotional measures, and especially tions. Freedom and generosity have even larger
negative emotions, are much less fully explained infuences on positive affect than on the ladder. 16
by the six variables than are life evaluations. But Negative affect is signifcantly reduced by social
the differences vary a lot from one circumstance support, freedom, and absence of corruption.
to another. Per-capita income and healthy life
expectancy have signifcant effects on life evalua- In the fourth column we re-estimate the life
tions, but not, in these national average data, on evaluation equation from column 1, adding both
either positive or negative affect. The situation positive and negative affect to partially
implechanges when we consider social variables. W ORLD HAPPINES S REPOR T 20 16 | UPD ATE
Technical Box 2: Detailed information about each of the predictors in Table 2.1
1. GDP per capita is in terms of Purchasing 4. Freedom to make life choices is the national
Power Parity (PPP) adjusted to constant 2011 average of binary responses to the GWP
international dollars, taken from the World question “Are you satisfed or dissatisfed
Development Indicators (WDI) released by with your freedom to choose what you do
the World Bank in December 2015. See the with your life?”
appendix for more details. GDP data for 2015
are not yet available, so we extend the GDP 5. Generosity is the residual of regressing the
time series from 2014 to 2015 using coun- national average of GWP responses to the
try-specifc forecasts of real GDP growth from question “Have you donated money to a
charthe OECD Economic Outlook No. 98 (Edition ity in the past month?” on GDP per capita.
2015/2) and World Bank’s Global Economic
Prospects (December 2014 release), after ad- 6. Perceptions of corruption are the average of
justment for population growth. The equa- binary answers to two GWP questions: “Is
tion uses the natural log of GDP per capita, corruption widespread throughout the
govsince that form fts the data signifcantly bet- ernment or not” and “Is corruption
wideter than does GDP per capita. spread within businesses or not?” Where data
for government corruption are missing, the
2. The time series of healthy life expectancy at perception of business corruption is used as
birth are constructed based on data from the the overall corruption-perception measure.
World Health Organization (WHO) and the
World Development Indicators (WDI). WHO 7. P ositive affect is defned as the average of
prepublishes the data on healthy life expectancy vious-day affect measures for happiness,
for the year 2012. The time series of life ex- laughter and enjoyment for GWP waves 3-7
pectancies, with no adjustment for health, (years 2008 to 2012, and some in 2013). It is
are available in WDI. We adopt the following defned as the average of laughter and
enjoystrategy to construct the time series of healthy ment for other waves where the happiness
life expectancy at birth: frst we generate the question was not asked.
ratios of healthy life expectancy to life
expectancy in 2012 for countries with both data. 8. Negative affect is defned as the average of
We then apply the country-specifc ratios to previous-day affect measures for worry,
sadother years to generate the healthy life expec- ness and anger for all waves. See the
appentancy data. See the appendix for more details. dix for more details.
3. S ocial support (or having someone to count
on in times of trouble) is the national average
of the binary responses (either 0 or 1) to the
Gallup World Poll (GWP) question “If you
were in trouble, do you have relatives or
friends you can count on to help you
whenev17er you need them, or not?”