WHR19 - Ch2 VF PDF

Introduction The Evolution of World Happiness
In the first World Happiness Report we surveyed 2005-2018

a wide range of available data. The Gallup World In recent previous reports, we presented bar
Poll surveys covering 2005-2011 gave the widest charts showing for the world as a whole, and for
international coverage. Now, seven years later, each of 10 global regions, the distribution of
we have twice as many years of data from the answers to the Cantril ladder question asking
Gallup World Poll, giving us a sufficient time respondents to value their lives today on a 0 to
span to consider how our principal measures of 10 scale, with the worst possible life as a 0 and
happiness, and their main supporting factors, the best possible life as a 10. This gave us a
have evolved from 2005 through 2018. chance to compare happiness levels and inequality
The chapter therefore starts with a presentation in different parts of the world. Population- 12
of the evolution of annual data at the global and weighted average life evaluations differed
13
regional levels for three key happiness measures significantly among regions, being highest in
– life evaluations, positive affect, and negative North America and Oceania, followed by Western
affect over the whole course of the Gallup World Europe, Latin America and the Caribbean,
Poll from 2005 through 2018. For all our plots of Central and Eastern Europe, the Commonwealth
annual data, we combine the surveys in 2005 of Independent States, East Asia, Southeast Asia,
and 2006, because of the small number of the Middle East and North Africa, Sub-Saharan
countries in the first year.1 Africa and South Asia, in that order. We found
that well-being inequality, as measured by the
The title of this chapter is intentionally ambiguous, standard deviation of the distributions of individual
designed to document not just the year-to-year life evaluations, was lowest in Western Europe,
changes in happiness, but also to consider how North America and Oceania, and South Asia; and
happiness has been affected by changes in the greatest in Latin America, Sub-Saharan Africa,
quality of government. After our review of how and the Middle East and North Africa.2
world happiness has been changing since the
start of the Gallup World Poll, we turn to present This year we shift our focus from the levels and
our rankings and analysis of the 2016-2018 distribution of well-being to consider their
average data for our three measures of subjective evolution over the years since the start of the
well-being plus the six main variables we use to Gallup World Poll. We now have twice as many
explain their international differences. See years of coverage from the Gallup World Poll as
Technical Box 1 for the precise definitions of all were available for the first World Happiness
nine variables. Report in 2012. This gives us a better chance
to see emerging happiness trends from 2005
For our country-by-country analysis of changes, through 2018, and to investigate what may
we report changes from 2005-2008 to 2016-2018, have contributed to them.
grouping years together to provide samples of
sufficient size. We shall also provide estimates of First we shall show the population-weighted
the extent to which each of the six key explanatory trends3, based on annual samples for the world
variables contributed to the actual changes in life as a whole, and for ten component regions, for
evaluations from 2005-2008 to 2016-2018. each of our three main happiness measures: life
evaluations, positive affect, and negative affect.
We then complete the chapter with our latest As described in Technical Box 1, the life evaluation
evidence on the links between changes in the used is the Cantril Ladder, which asks survey
quality of government, by a variety of measures, respondents to place the status of their lives on
and changes in national average life evaluations a “ladder” scale ranging from 0 to 10, where 0
over the 2005-2018 span of years covered by the means the worst possible life and 10 the best
Gallup World Poll. possible life. Positive affect comprises the
average frequency of happiness, laughter and
enjoyment on the previous day, and negative
affect comprises the average frequency of worry,
sadness and anger on the previous day. The
affect measures thus lie between 0 and 1.
World Happiness Report 2019
The three panels of Figure 2.1 show the global favoured the largest countries, as confirmed by
and regional trajectories for life evaluations, the third line, which shows a population-weighted
positive affect, and negative affect. The whiskers average for all countries in the world except the
on the lines in all figures indicate 95% confidence five countries with the largest populations – China,
intervals for the estimated means. The first panel India, Indonesia, the United States and Russia.4
shows the evolution of life evaluations measured The individual trajectories for these largest
three different ways. Among the three lines, two countries are shown in Figure 1 of Statistical
lines cover the whole world population, with one Appendix 1, while their changes from 2005-2008
of the two weighting the country averages by to 2016-2018 are shown later in this chapter, in
each country’s share of the world population, Figure 2.8. Even with the largest countries
and the other being an unweighted average of removed, the population-weighted average does
the individual national averages. The unweighted not rise as fast as the unweighted average,
average is always above the weighted average, suggesting that smaller countries have had
especially after 2015, when the weighted average greater happiness growth since 2015 than have
starts to drop significantly, while the unweighted the larger countries.
average starts to rise equally sharply. This
suggests that the recent trends have not
Cantril Ladder Cantril Ladder Positive Affect
5.7 5.7 0.78
5.6 5.6
0.76
5.5 Figure
5.5 2.1: World Dynamics of Happiness
5.4 5.4 0.74
Cantril Ladder
Cantril Ladder Cantril Ladder
Positive Affect
Positive Affect Positive Af
5.3 5.3 Cantril Ladder 0.72 Positive Affect
.7 5.7 5.7 5.7 0.78 0.78 0.78
5.2 5.2 0.78
.6 5.6 5.1
5.6 5.6 0.70
5.1 0.76 0.76 0.76
0.76
.5
5.0 5.5 5.0
5.5 5.5 0.68
0.74 0.74 0.74
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
.4 5.4 5.4 5.4
2006 2006
2007
2007
2008
2009
2010
2008
2011
2012
2013
2014
2009 2015
2016
2017
2010 2018
2011
2012
2013
2014
2015
2016
2017
0.74
.3 5.3 5.3 5.3 0.72 0.72 0.72

Population weighted
Population weighted 0.72
Popupation weighted
.2 5.2 5.2 5.2
0.70 0.70 0.70 Non‐population weighted
.1 5.1 5.1 5.1 Population weighted (excluding top 5
0.70
Population weighted (excluding top 5
.0 5.0 5.0 5.0 largest countries)
largest countries) 0.68 0.68 0.68
0.68
20092006 2006
2007
2006
20102007
2008
2007
2009
2008
2010
2009
2011
2010
2012
2008
2011
2013
2012
2014
2013
2015
2014
2009 2016
2015
2017
20122010
2016
2018
2017
2018
20132011
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2016
2017
2018 2006
2006
2007
2006
2008
2007
2009
2008
2010
2009
2011
2007
2010
2012
2011
2013
2012
2014
2013
2015
2014
2008
2016
2015
2017
2016
2018
2017
2018
2011
Non‐population weighted
Population weighted
Population weighted
Population weighted
Negative Affect Negative Affect Popupation w
Popupation weighted
Popupation weighted
Negative Affect
0.32 0.32 0.32 Non‐population weighted
Non‐populatio
g top 5 Population weighted (excluding top 5
Population weighted (exc
Population weighted (excluding top 5
0.30 0.30 0.30 largest countries)
largest countries)
largest countries)
0.28 0.28 0.28Non‐population weighted
Non‐population weighte
0.26 0.26 0.26
Negative Affect
Negative Affect Negative Affec
0.24 0.24 0.24
.32 0.32 0.32
 Population weighted
0.22 0.22 0.22
.30 0.30 0.30  Population
weighted (excluding top 5 largest countries)
0.20 0.20 0.20  Non-population weighted

.28 0.28 0.28
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
0.18 0.18
.26 0.26 0.26
20062006
2007
2008
20072009
2010
2011
2008
2012
2013
2014
2015
20092016
2017
2018
2010
2011
2012
2013
2014
2015
2016
2017
.24
0.24 0.24
Popupation weighted
Popupation weighted
.22
0.22 0.22
.20
0.20 0.20
The second panel of Figure 2.1 shows positive with about 80% of its population in the United
affect over the same period as used in the first States. The weighted average, heavily influenced
panel. There is no significant trend in either the by the U.S. experience, has fallen more than 0.4
weighted or unweighted series. The population- points from its pre-crisis peak to 2018, about on
weighted series show slightly but significantly a par with Western Europe. The lower line shows
more positive affect than does the unweighted that average happiness in Latin America and the
series, showing that positive affect is on average Caribbean rose without much pause until a peak
higher in the larger countries. in 2013, with a continuing decline since then.
The third panel of Figure 2.1 shows negative The third panel shows quite different evolutions
affect, which follows a quite different path from of life evaluations in the three parts of Asia, with
positive affect. The population-weighted world South Asia showing a drop of a full point, from 14
frequency of negative affect in 2005-2006 is 5.1 to 4.1 on the 0 to 10 scale, driven mainly by
about one-third of the frequency of positive the experience of India, given its dominant share 15
affect. Negative affect is lower for the weighted of South Asian population. Southeast Asia and
series, just as positive affect is greater. Both the East Asia, in contrast, have had generally rising
weighted and unweighted series show significant life evaluations over the period. Southeast and
upward trends in negative affect starting in 2010 South Asia had the same average life evaluations
or 2011. The global weighted measure of negative in 2005-2006, but the gap between them was
affect rises by more than one-quarter from 2010 up to 1.3 points by 2018. Happiness in East Asia
to 2018, from a frequency of 22% to 28%. This was worst hit in the economic crisis years, but
global total, striking as it is, masks a great deal of has since posted a larger overall gain than
difference among global regions, as will be Southeast Asia to end the period at similar levels.
shown later in Figure 2.4.
Finally, the fourth panel of Figure 2.2 contains
The four panels of Figure 2.2 show the evolution the Middle East and North Africa (MENA) and
of life evaluations in ten global regions, divided Sub-Saharan Africa (SSA), with MENA dropping
into four continental groupings.5 In each case the fairly steadily, and SSA with no overall trend. In
averages are adjusted for sampling and population all regions there is a variety of country experiences
weights. The first panel has three lines, one each underlying the averages reported in Figure 2.2.
for Western Europe, Central and Eastern Europe, The country-by-country data are reported in the
and the Commonwealth of Independent States on-line statistical data, and the country changes
(CIS). All three groups of countries show average from 2005-2008 to 2016-2018 shown later in
life evaluations that fell in the wake of the 2007- Figure 2.8 will help to reveal the national sources
2008 financial crash, with the falls being greatest of the regional trends.
in Western Europe, then in the CIS, with only a
The four panels of Figures 2.3 and 2.4 have the
slight drop in Central and Eastern Europe. The
same structure as Figure 2.2, with life evaluations
post-crash happiness recovery started first in the
being replaced by positive affect in Figure 2.3
CIS, then in Central and Eastern Europe, while in
and by negative affect in Figure 2.4. Figure 2.3
Western Europe average life evaluations only
shows that positive affect is generally falling in
started recovering in 2015. CIS evaluations rose
Western Europe, and falling and then rising in
almost to the level of those in Central and
both Central and Eastern Europe and the CIS,
Eastern Europe by 2014, but have since fallen,
achieving its highest levels at the end of the
while those in Central and Eastern Europe have
period. This pattern of partial convergence of
continued to rise, parallelling the post-2015 rise
positive affect between the two parts of Europe
in Western Europe. The overall pattern is one of
leaves positive affect still significantly more
happiness convergence among the three parts of
frequent in Western Europe. Within the Americas,
Europe, but with a recent large gap opening up
the incidence of positive affect is generally
between Central and Eastern Europe and the CIS.
falling, at about the same rates in both the
The second panel of Figure 2.2 covers the NA-ANZ region (with most of the population
Americas. The upper line shows the North weight being on the United States), and in Latin
America+ANZ country grouping comprising the America. Positive affect is fairly stable and at
United States, Canada, Australia and New Zealand, similar levels in East and Southeast Asia, while
its
its pre-crisis
its pre-crisis
peakpeak
to 2018,
to 2018,
about
about
on aon
para with
pre-crisi par with
Western
itsWestern
Europe.
Europe.
The The
lower
lower
linepre-
line
showsshows
th that
that
Western
its that
average
itsits
World Happiness average
happiness
pre-crisis
pre-crisis
Report 2019 happiness
peak
peak in2018,
to to Latin
2018,inabout
Latin
America
onAmerica
about
pre-cris aand
aonpar parthe
and
with Caribbean
theWestern
averag
that
with Caribbean
Western rose
Europe. rose
without
Europe.
The without
Themuch
lower linemuch
lower pause
shows pause
shows av
lineEur
until
until
the
that until
a peak
that
that a peak
inCaribbean
average
average2013,
in 2013,
within
happiness
happinesswith
aLatin
continuing
in a continuing
Latin decline
America
Americadecline
and since
and
the since
untilthen.
a
Caribbean
averag
the then.
Caribbean peak
rose without
rose without
much much
pause pause a
ne since
until
until
untila apeak
peakinin
2013,
2013,
with
with
a continuing
a continuing
decline
decline
sincesince
then.
then.then.
a pea
Figure Figure
2.2: Dynamics
Figure
Figure of Ladder
2.2 Dynamics
2.2 Dynamics ofin
of Ladder10inRegions
Ladder 10inRegions
10 Regions
Figure
2.2 D
s Figure
Figure
Figure2.2
2.2Dynamics
Dynamicsof of
Ladder
Ladder
in 10
in Regions
10 Regions 2.2
Europe
Europe The Americas and ANZ
The Americas and ANZ
Europe
0 8.0 8.0 Europe
8.0 Europe 8.0 8.0 The Americas and ANZ
8.0 8.0

5 7.5 8.0
8.0
7.5 8.0
7.5 8.0 7.5
7.5 8.0
7.5 Western
Western 7.5

0 7.5
7.5 7.5 Europe
7.0 7.0 7.0 Europe
Western
Western 7.5 7.0
7.0 7.5
7.0 7.0
Europe
Europe
5 6.5 7.0
7.0 7.0
6.5
6.5
6.5 7.0 7.0
6.5
6.5
6.5
6.5
6.5 6.5 6.5 6.5
0 6.0 6.0
6.0
6.0 Central and
Central and 6.0
6.0
6.0 North America
North America
6.0
6.0 6.0 Eastern Europe
Central and
Central and
Eastern Europe6.0 6.0 North America
North America
and ANZ
and ANZ
5 5.5 5.5
5.5
5.5 5.5
5.5
5.5 and ANZ and ANZ
5.5
5.5 5.5 Eastern Europe
Eastern Europe5.5 5.5 Latin America
Latin America
0 5.0 5.0 5.0 5.0 5.0 Latin America
Latin America

5.0
5.0
5.0 5.0
5.0
5.0 5.0 and Caribbean
and Caribbean
5 4.5 4.5 4.5 Commonwealt and Caribbean
and Caribbean
4.5 Commonwealt4.5
4.5
4.5
4.5
4.5 4.5 Commonwealt
Commonwealt4.5 4.5
0 4.0 4.0 4.0 h of h of 4.0 4.0
4.0
4.0
4.0 h of Independent
h of
4.0 Independent 4.0
4.0 4.0
2006
2008
2010
2012
2014
2016
2018
2006
2008
2010
2012
2014
2016
2018
5 Independent
3.5 3.5 3.5 States Independent 3.5 3.5
3.5
3.5 3.5 States 3.5 3.5
States
States
 Western Europe  North America and ANZ
 Central
and Eastern Europe  Latin America and Caribbean
 Commonwealth of Independent States
Asia Asia
Asia
Asia Africa and Middle East
Africa and Middle East
0
8.0 8.0
8.0 8.0
8.0 8.0 8.0
8.0 8.0
8.0
8.0 8.0
5
7.5 7.5
7.5 7.5
7.5 7.5 7.5
7.5 7.5
7.5
7.5 7.5
0
7.0 7.0
7.0
7.0
7.0
7.0
7.0 7.0
7.0
7.0
7.0
7.0
5
6.5 6.5
6.5
6.5 6.5
6.5 6.5
6.5 6.5
6.5
6.5 6.5 Middle East
Middle East
Middle East
Middle East

0 6.0 6.0
6.0
6.0 6.0 East Asia
6.0 6.0 East Asia
East Asia 6.0
6.0
6.0
6.0
6.0 and North
and North
and North
and North
Africa Africa
AfricaAfrica

5 5.5 5.5
5.5
5.5 5.5 Southeast Asia 5.5
Southeast Asia5.5
5.5 5.5 Southeast Asia
5.5
5.5
Southeast Asia 5.5
Sub‐Saharan
Sub‐Saharan
Sub‐Saharan
Sub‐Saharan

0 5.0 5.0
5.0
5.0 5.0
5.0 5.0 South Asia
South Asia 5.0
5.0 5.0
5.0 5.0
South Asia
South Asia Africa Africa
AfricaAfrica

5 4.5 4.5
4.5
4.5 4.5
4.5 4.5
4.5
4.5
4.5
4.5
4.5
4.0 4.0 4.0
0 4.0 4.0
4.0
4.0
4.0 4.0 4.0
4.0 4.0
2006
2008
2010
2012
2014
2016
2018
2006
2008
2010
2012
2014
2016
2018
3.5
3.5 3.5 3.5 3.5
5 3.5 3.5 3.5 3.5 3.5
 East Asia  Middle East and North Africa
 Southeast
Asia  Sub-Saharan Africa
 South Asia
ions
The
The
Thethird
thirdpanel
panel
shows
ofshows
quite
quite
different
different
evolutions
evolutions
life ofthird
life
ofevaluations
life evaluations
in thein
three
the parts
threeeva
ofparts of
The
The The
thirdthird
panelpanel
showsshows
quitequite
different
different
evolutions
evolutions
of life
third
The of evaluations
life evaluations
in the
inthree
the three
partsparts
of of th
p
ull point,
Asia,
Asia,
Asia,with
withSouth
SouthAsia
Asia
showing
showing
a drop
a drop
of aof
full
a full
point,
point,
fromfrom
5.1
fromto5.1
4.1
with
toon4.1
theon
0 to
the100scale,
to 10 scale,
Asia,
Asia,
Asia, withwith
SouthSouth
AsiaAsia
showing
showing
a drop
a drop
of a of
Asia thefulla point,
full point,
Asia, fromfrom
5.1 to
with5.14.1
to on
4.1the
on0the
to 010toscale,
10 scale, w
S
starting lower
ven and
driven falling
driven
driven
its significantly
mainly
mainly
bybythethe in Southofdominan
experience
experience India,
of India,
given itsfrequency
given dominant of negative
its dominant
shareshare ofaffect
of South South
Asianrises
ma most
Asian
driven
Asia. There are driven
driven
no mainly
mainly
by trends
significant the
byexperience
theinexperience
of India,
positive of India,
givengiven
sharply its
in dominant
its dominant
driven
Southeast share
share
Asia,of South
andofbySouth
main Asian
only Asian
slightly
contrast, population
population.
population.
affect in Sub-Saharan Southeast
Southeast
Africa, while inAsia
Asia
andand
MENA, East
it East
Asia,
Asia,
inin
less contrast,
in contrast,
South havehave
Asia, had generally
whilehad
have
generally
falling rising
in Eastrising
life
Asia life
until
population.
starts lower and population.
population.
follows Southeast
Southeast
Asia
a declining Asia
and East
trend. and East
Asia,
Asia,
in contrast,
2014 inpopula
andcontrast,
thenhave have
hadthereafter.
rising generally
had generally
rising
In therising
life life
Middle
East and North Africa, the frequency at first falls
Figure 2.4 shows that negative affect is generally
6 and 6 then6 rises, but within a narrow range. The
increasing in Western Europe, generally lower
6 6increases in the frequency of negative
biggest
and falling since 2012 in Central and Eastern
affect are found in Sub-Saharan Africa, with the
Europe, and also falling in the CIS until 2015, but
2018 frequency greater by half than in 2010. Thus
rising thereafter. Negative affect thus shows
all global regions except for Central and Eastern
divergence rather than the convergence within
Europe have had significantly increasing negative
Europe seen for life evaluations and positive
affect in recent years, with some variations
affect. There is a continuing post-crisis increase
among regions in starting dates for the increases.
in the incidence of negative affect in Latin
America as well as in the NA-ANZ region. Within
Figure 2.3: Dynamics of Positive Affect in 10 Regions
Europe and CIS
Europe and CIS The Americas and ANZ
90 0.90 0.90 0.90 0.90 0.90
Europe and CIS
0.90 0.90
85 0.85 0.85

0.85
0.85
Western Western 0.85 0.85
0.85
0 0.90 0.90 0.90 Europe Europe 0.90 0.90
80 0.80 0.80
0.80
0.80 0.80
0.80
0.80
5 0.85 0.85 0.85 Western 0.85
Western 0.85
75 0.75 0.75
0.75
0.75 Europe Europe 0.75
0.75
0.75
0 0.80 0.80 0.80 Central and 0.80
Central and 0.80 North America
North Ame
70 0.70 0.70 0.70 0.70 0.70 and ANZ and ANZ
e 0.70 Eastern Europe
Eastern Europe0.70
5 0.75 0.75 0.75 0.75 0.75 16
65 0.65 0.65 0.65 0.65 0.65 Latin America
Latin Ameri
North America
North Ame
0.65 Central and
Central and 0.65
0 0.70 0.70 0.70 0.70 0.70 and Caribbean
and ANZ and Caribbe
and ANZ
e
60 0.60 0.60 0.60 Eastern Europe
Eastern Europe
0.60 0.60 17
0.60 0.60
t
5 0.65 0.65 0.65 Commonwealt
Commonwealt 0.65 0.65 Latin America
Latin Ameri
55 0.55 0.55
0.55
h of
0.55 h of 0.55
0.55
0.55 and Caribbean
and Caribbe
0 0.60 0.60 0.60 Independent
Independent 0.60 0.60
50 0.50 0.50
0.50 0.50
Commonwealt
Commonwealt
States States 0.50
0.50 0.50
5 0.55 0.55 0.55 h of h of 0.55 0.55
2006
2008
2010
2012
2014
2016
2018
2006
2008
2010
2012
2014
2016
2018
Independent
Independent
0 0.50 0.50 0.50 States States 0.50 0.50
Asia
 Central and Eastern Europe Asia Africa and Middle East
 Latin America and Caribbean
90 0.90 0.90 0.90 0.90 0.90
Asia Asia Africa and Middle East
85 0.85 0.85 0.85 Asia 0.85 0.85
0 0.90 0.90 0.90
0.90
0.90
0.90
0.90
80 0.80 0.80 0.80 0.80 0.80
5 0.85 0.85 0.85
0.85
0.85
0.85
0.85
75 0.75 0.75 0.75 0.75 0.75 Middle East
Middle E
0 0.80 0.80 0.80
0.80 East AsiaEast Asia 0.80
0.80
0.80 and North
and Nort
70 0.70 0.70 0.70 0.70 0.70 Africa Africa
5
a 0.75 0.75 0.75 Southeast Asia 0.75
Southeast Asia 0.75 Middle East Middle E
0.75 0.75
65 0.65 0.65 0.65 East Asia 0.65
East Asia 0.65 Sub‐Saharan
and North Sub‐Saha
and Nort
0 0.70 0.70 0.70 South Asia
South Asia 0.70 0.70 Africa Africa
Africa
0.70 0.70
60
a 0.60 0.60 0.60 Southeast Asia 0.60
Southeast Asia 0.60
5 0.65 0.65 0.65
0.65
0.65
0.65
0.65 Sub‐Saharan
Sub‐Saha
55 0.55 0.55 0.55 South Asia 0.55
South Asia 0.55 Africa Africa
0 0.60 0.60 0.60
0.60
0.60
0.60
0.60
50 0.50 0.50 0.50 0.50 0.50
5 0.55 0.55 0.55
0.55
0.55
0.55
0.55
0 0.50 0.50 0.50
0.50
0.50
0.50
0.50
2006
2008
2010
2012
2014
2016
2018
2006
2008
2010
2012
2014
2016
2018
erally
Figure
Figure Figure
2.4 shows
 East 2.4 shows
Asia thatincreasing
negative
that negative
affect is
affect
generally
is generally
increasing
increasing
 Middle in and
East Western
2.4 in Africa
North Western
Europe,
Europe,
generally
generally
 Southeast Asia  Sub-Saharan Africa
Eastern
lower
lower and
lower falling
 Southand
Asia falling
since 2012
sincein2012Central
in Central
Europe,
and Eastern
and Eastern
Europe,
and Europe,
and alsoandfalling
also falling
in the CISin the CIS f
erally Figure
Figure 2.4 shows
Figure 2.4 shows that
increasing
negative affectaffect
that negative is generally increasing
is generally in Western
increasing Europe, Europe,
in Western generallygenerally
2
ffectuntil
until 2015,
untilthus
but
2015, rising
but thereafter.
rising thereafter.
Negative Negative
affect
2015,
thus
shows
affectshows
thus shows
divergence
divergence
rather than
rather thethan the
Eastern
lowerlower and and
lower falling sincesince
falling 2012 2012
in Central Europe,
and Eastern
in Central Europe,Europe,
and Eastern and alsoand
falling
alsoinfalling
anthe CIS in the CIS
aluations
convergenc
convergence
convergence
within within
EuropeEuropeseen for seen
lifefor evaluations
life
andevaluations
and positive
and positive
affect. affect.
Therepos is
There
a is a
ffectuntiluntiluntil
2015,
thus but rising
2015, thereafter.
but rising Negative
thereafter. affect shows
Negative thus
affectshows
thus divergence
shows
201 rather than
divergence the than the
rather
nce continuing
continuing
of continuing
post-crisis
post-crisis
increase
negative
increase
in the incidence
in the incidenceof negative
of negative
affect in
affect
LatininAmerica
Latin Americaas as a
aluations convergence
converge convergence within Europe
within seen for
Europe seen lifeforevaluations
and
life and positive
evaluations affect. There
and positive affect.is There
pos
a is a
a thewell
well aswell
infrequency
the
as NA-ANZ
in the NA-ANZ region.region.
WithinWithin Asia as the
Asia
frequency
the frequency
of negative
of negative
in
affect rises
affectmostrises mostot
nce of
continuing
continui continuing post-crisis negative
increase
post-crisis in thein
increase incidence of negative
the incidence affect inaffect
of negative LatininAmerica as
Latin America as a
htly sharply
sharply
less sharply
in Southeast
in Southeast
Asia, and Asia,byandonlyin
byslightly
only slightly
less in less
South
South
in Asia,
Southwhile
Asia,infalling
while falling
in Eastin East S
thewell wellwell
frequency
as inasthein NA-ANZ
the NA-ANZ region. Within
region. Asia the
Within frequency
Asia of negative
the frequency
as affect rises
of negative most
affect rises most oi
n the
Asia
Asia until
Asia2014until
Middle
and
2014 thenand rising
then thereafter.
rising thereafter.In the
untilMiddle
In the Middle
East andEast
North
and Africa,
North
Eas Africa,
the the 2
htly less
sharply
sharply sharplyin Southeast
in SoutheastAsia,Asia,
and by andonly
inbyslightly less in less
only slightly South
South
in Asia,
Southwhile
Asia,falling
whileinfalling
East in East i
ithinfrequency
frequency
frequency
at
a first atfalls
firstand
falls
narrow
thenandrises,
thenbut rises,within
but within
a narrow a narrow
range. range.
The biggest
The biggest
increases increases
rain in a
n the
Asia AsiaAsia
untiluntil
Middle
20142014 and then risingrising
and then thereafter. In the In
thereafter. Middle East and
the Middle unt North
East andAfrica,
NorthEas the
Africa, the
n Sub-Saharan
the
the frequency
the frequency
of negative
of negative
affectfrequen
are
affectfound are infound
Sub-Saharan
in Sub-Saharan
Africa,Africa,
with the with
2018 the 2018 A
ithin frequency
frequenc a
frequency at first fallsnarrow
at first and then
falls and rises, but within
then rises, a narrow
but within range. The
a narrow range.biggest
The increases in ra
biggest increases in
n 8
Sub-Saharan
the the
the frequency of negative
frequency affectaffect
of negative are found
are in8 Sub-Saharan
frequ
found Africa, Africa,
in8Sub-Saharan with thewith
2018the 2018 A
8 8 8
l global
frequency
frequency
frequency
greatergreater
by halfby
than
half
inthan
2010.
in Thus
2010.all
Thus
global
all global
regionsregions
exceptregi
except
for Central
for Central
and and g
ng Eastern
EasternEastern
negative
EuropeEurope
have hadhavesignificantly
had significantly
increasing
increasing
negative negative
affect in
affect
recent
in recent
years,
Eur
with
years, with a
llfrequency
global
frequency
frequencygreater
greater
by half
by half
thanthan
in 2010.
in 2010.
ThusThus
all global
all global
regionsregions
exceptregio
except
for Central
for Central
and and g
s for
somesome variations
some variations
amongamong
regionsregions
in
thestarting
in starting
dates fordates
the for
increases.
variat
the increases. incr
ingEastern Eastern
Eastern
Europe
negative
Europehave
have
hadhad
significantly
significantly
increasing
increasing
negativenegative
affect affect
in recent
in recent
years, Eur
with
years, with a
Figure 2.4: Dynamics of Negative Affect in 10 Regions
es some
Regions for
Figure some
Figure some
2.4variations
variations
Figure
Dynamics among
among
2.4 Dynamics
of regions
ofregions
Negative in
the
starting
Negativein starting
Affect in dates
Affect dates
for10the
10 Regions
in for
variat
increases.
the increases.
Regions incre
2.4 D
EuropeEurope The Americas and ANZ
Regions
Figure
Figure
Figure
2.42.4
Dynamics
Dynamics
Europe of Negative
of Negative
Affect
Affect
in 10in
Regions
10 Regions 2.4
The Americas and ANZ D
0.45
0.45 0.45
0.45
0.45
0.45
0.45
Europe
0.40
0.40 0.40 Western Western 0.40 0.40
0.40 0.40
0.45
0.45
0.45 Europe Europe 0.45 0.45
0.35
0.35 0.35
0.35
0.35
0.35
0.35
0.40
0.40
0.40 Western
Western 0.40 0.40
0.30
0.30 0.30 EuropeEurope 0.30 0.30 North America
North America
0.30 Central and
Central and 0.30
0.35
0.35
0.35 0.35 0.35 and ANZ and ANZ
Eastern Europe
Eastern Europe
0.25
0.25 0.25
0.25
0.25
0.25
0.25
0.30
0.30
0.30 0.30 0.30 Latin America
Latin America
North America
North America
Central and
Central and
0.20
0.20 0.20 0.20 0.20 and Caribbean
and ANZ and Caribbean
and ANZ
0.20 Eastern Europe
Eastern Europe 0.20
0.25
0.25
0.25 Commonwealt
Commonwealt 0.25 0.25
0.15
0.15 0.15 h of h of 0.15 0.15 Latin America
Latin America
0.15 0.15
0.20
0.20
0.20 Independent
Independent 0.20 0.20 and Caribbean
and Caribbean
0.10
0.10 0.10
0.10 Commonwealt
States Commonwealt
States 0.10
0.10
0.10
0.15
0.15
0.15 0.15 0.15
2006
2008
2010
2012
2014
2016
2018
2006
2008
2010
2012
2014
2016
2018
h of h of
Independent
Independent
0.10
0.10
0.10 StatesStates 0.10 0.10
 Central
Asia
and Eastern Europe Asia Africa and Middle East
 Latin America and Caribbean
0.45
0.45 0.45 0.45 0.45
AsiaAsia Africa and Middle East
0.40
0.40 0.40 Asia 0.40 0.40
0.45
0.45
0.45
0.45
0.45
0.45
0.45
0.35
0.35 0.35 0.35 0.35
0.40
0.40
0.40
0.40
0.40
0.40
0.40 Middle East
Middle East
0.30
0.30 0.30 0.30 0.30 and North and North
East AsiaEast Asia
0.35
0.35
0.35
0.35
0.35
0.35
0.35 Africa Africa
0.25
0.25 0.25 Southeast Asia 0.25
Southeast Asia 0.25 Middle East
Middle East
0.30
0.30
0.30 0.30 0.30 Sub‐Saharan
and North Sub‐Saharan
and North
0.30 East Asia
East Asia
South Asia 0.30
South Asia 0.20
0.20
0.20 0.20 0.20 Africa Africa
Africa Africa
0.25
0.25
0.25
0.25
Southeast Asia 0.25
Southeast Asia 0.25
0.25
0.15
0.15 0.15 0.15 0.15 Sub‐Saharan
Sub‐Saharan
0.20
0.20
0.20 South Asia
South Asia 0.20 0.20 Africa Africa
0.20 0.20
0.10
0.10 0.10 0.10 0.10
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.10
0.10
0.10
0.10
0.10
0.10
0.10
2006
2008
2010
2012
2014
2016
2018
2006
2008
2010
2012
2014
2016
2018
y The
The Evolution
The
 East Evolution
Asia
 Southeast
Asia
of Happiness
of Happiness
Inequality
Inequality
Evol
 Middle East and North Africa
 Sub-Saharan Africa
 South Asia
esIn in
In this In
section
this section
we focus
we
this
our
focus ourthe
attention
attention
on changes
on changes
in the distribution
in the distribution
secti dist
of happiness.
of happiness.
There There
tyThe
The
TheEvolution
Evolutionof Happiness
of Happiness
Inequality
Inequality
Evol
it is
are
are at least
are attwo
least two importan
reasons reasons
for us to
forat
dousthis.
to doFirst,
this.itFirst,
is important
it is important
leastto consider
to consider
not justnotaverage
just average
gesIn In In
this
this
section
insectionwe wefocus
this
focus
our our
attention
attention
the on changes
on changes
in thein distribution
the distribution
secti
of happiness.
distr
of happiness.
There There
howhappiness
happiness
happiness
in a community
in a community it
or country,
or country,
but alsobut
how also is
it how
is distributed.
it is distributed.
Second, it isdis
Second, done
it istodone to i
itare is
areare
at least
at least
twotwo reasons
reasons
importan
for us
forat
tousdotothis.
do this.
First,First,
it is important
it is important
leastto consider
to consider
not justnot
average
just average
sider
encourage
encourage
encourage
those interested
those interested happin
in inequality
in inequality
to consider
to consider
happiness happiness
inequality
inequality
as a useful
as a useful t
how
happiness
happiness
happiness in aincommunity
a community or country,
oritcountry,
but also
but how
also ithow
is
is distributed.
it is distributed.
Second,
Second,
itdist
is done
it is
to done to i
nsider
encourage
encourage
encourage
those
those
interested
interested
in inequality
in inequality
happine
to consider
9 to consider
9happiness
happiness
inequality
inequality
as a useful
as a useful t
9 9
The Evolution of Happiness Inequality Figure 2.5 shows the evolution of global inequality
of happiness, as measured by the standard
In this section we focus our attention on changes
deviation of the distribution of the individual life
in the distribution of happiness. There are at least
evaluations on the 0 to 10 scale, from 2005-2006
two reasons for us to do this. First, it is important
to 2018. The upper line illustrates the trend of
to consider not just average happiness in a
overall inequality, showing a clear increase since
community or country, but also how it is
2007. We further decompose overall inequality
distributed. Second, it is done to encourage
into two components: one for within-country
those interested in inequality to consider
inequality, and another for between-country
happiness inequality as a useful umbrella
inequality. The figure shows that inequality within
measure. Most studies of inequality have focused
countries follows the same increasing trend as
on inequality in the distribution of income and 18
overall inequality, while between-country
wealth,6 while in Chapter 2 of World Happiness
inequality has increased only slightly. In summary,
Report 2016 Update we argued that just as 19
global happiness inequality, measured by the
income is too limited an indicator for the overall
standard deviation of Cantril Ladder, has been
quality of life, income inequality is too limited
increasing, driven mainly by increasing happiness
a measure of overall inequality.7 For example,
inequality within countries.
inequalities in the distribution of health8 have
effects on life satisfaction above and beyond those Figure 2.6 shows that the inequality of happiness
flowing through their effects on income. We and has evolved quite differently in the ten global
others have found that the effects of happiness regions. The inequality of happiness rose between
equality are often larger and more systematic than 2006 and 2012 in Western Europe, and has been
those of income inequality. For example, social falling steadily since, while in Central and Eastern
trust, often found to be lower where income Europe it has followed a similar path but starting
inequality is greater, is even more closely connected from a higher starting point and falling faster.
to the inequality of subjective well-being.9 Inequality in the CIS region follows somewhat
the reverse pattern, being stable at first and
Figure 2.5: Dynamics of Inequality of Ladder (Standard Deviation)
Inequality of World Happiness
2.8
2.4
2.0
1.6
1.2
0.8
0.4
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
 SD
 Within
SD
 Between SD
0.4 0.4
0.4

0.0 0.0
0.0
2012 2006 2007
2006 2008
2007 2009
2008
2013
2010
2009 2011
2010 2012
2011 2013
2012
2006
2014
2013 2015
2014 2016
2015 2017
2016
20142018
2017 2018 2
SD SD SDWithin SD
Within SDBetween SD
Betwee
Between SD
Figure 2.6: Dynamics of Inequality of Ladder in 10 Regions

r in
Figure 2.610
Figure Figure
Dynamics
2.6 Dynamics Regions
of Inequality
of Inequality
of Ladder
of Ladder
in 10 2.6
Regions
in 10 Regions Dyn
EuropeEurope The Americas and ANZ
0 3.00 3.00
3.00
3.00 3.00
3.00
3.00
Western Western
0 2.50 2.50
2.50
2.50
Europe Europe 2.50
2.50
2.50
0 2.00 2.00 2.00 2.00 2.00

2.00
Central and
Central and
2.00 North America
North America
Eastern Europe
Eastern Europe and ANZ and ANZ
0 1.50 1.50
1.50
1.50 1.50
1.50
1.50 Latin America
Latin America
and Caribbean
and Caribbean
0 1.00 1.00
1.00
Commonwealt
1.00 Commonwealt
1.00
1.00
1.00
h of h of
Independent
Independent
0 0.50 0.50
0.50
0.50
States States 0.50
0.50
0.50
2006
2008
2010
2012
2014
2016
2018
2006
2008
2010
2012
2014
2016
2018
 Central
and Eastern Europe  Latin America and Caribbean
Asia Asia Africa and Middle East

Africa and Middle East A
0 3.00 3.00
3.00
3.00 3.00
3.00
3.00
11
0 2.50 2.50 2.50 11 11 2.50
2.50
2.50 2.50
0 2.00 2.00 2.00 2.00 2.00 Middle East

Middle East
2.00
East Asia
East Asia
2.00
and North and North
Africa Africa
a0 1.50 1.50 Southeast Asia
1.50 Southeast Asia
1.50 1.50
1.50 1.50 Sub‐Saharan
Sub‐Saharan
South Asia
South Asia Africa Africa
0 1.00 1.00
1.00
1.00 1.00
1.00
1.00
0 0.50 0.50
0.50
0.50 0.50
0.50
0.50
2006
2008
2010
2012
2014
2016
2018
2006
2008
2010
2012
2014
2016
2018
 East Asia  Middle East and North Africa

 Southeast
Asia  Sub-Saharan Africa
iness
Figure
Figure
Figure
2.6 shows
2.6 shows
that
hasthe
thatinequality
the inequality
of happiness
of happiness
 South Asia
has
2.6
evolved
evolved
has evolved
quite differently
quite differently
in
show
the ten
in the ten
rose
global
global
global
regions.
regions.
The
between
The
inequality
inequality
of happiness
of happiness
roseregions.
between
rose between
2006 and
2006
2012
andin2012
Western
in Western
200
while
Europe,
Europe,
Europe,
and and
has been
has been
falling
in falling
steadily
steadily
since,since,
while
Central
while
in Central
in Central
and
and Eastern
and Eastern
EuropeEurope
it has it hashas
a
rising since 2013. In Latin America, inequality was Ranking of Happiness by Country
igher
steadyfollowed
followed
until 2014followed
and ahas
similar
arisen
similar
pathstarting
path
butwhile
since, starting
but rising
starting
from from
a higher
a higher
starting
starting
point and
point
falling
a and falling
faster. faster.
simi poin
Now we turn to consider life evaluations covering
until 2010
hat in Inequality
the
Inequality
theUS-dominated
Inequality
in the theNA+ANZ
in CIS CIS
region region
region
follows
reverse
follows
somewhat
somewhat
the reverse
the reverse
pattern,pattern,
being stable
being
in
atstable
first at firstpat
th
the 2016-2018 period, and to present our
and being fairly constant since. Inequality in
quality and andand
rising
rising
sincesince
2013. 2013.
In Latin
rising
In Latin
America,
was
America,
inequality
annual inequality
was steady
country was steady
untilThese
rankings. 2014
until
steady
and
2014
since
hasand
rankings risen
has risen
are
Southeast Asia has been rising throughout the
minated
periodsince,
since since,
since,
2010, while
while
while rising
in therising
until
rest until
2010
of Asia2010
inrising
NA+ANZ the accompanied
theinUS-dominated
US-dominated
while NA+ANZby ourregion
NA+ANZ latestregion
attempts
and being to show
andfairly
being fairly risi
how six key variables contribute to explaining
much
has less. Inequality
constant constant in
constant
beenSub-Saharan
since.
since.
Inequality Africa
Inequality has
in Southeast
in Southeast
Asia has
Asiabeenhas
rising
rising
been rising
throughout
since.
throughout
the period
the since
period since th
the full sample of national annual average scores
risen on the steep post-2010 path similar to that
e ss. 2010, 2010,
2010,
while
Inequality
while
in the
inrest
the of
rest
Asia
of Asia
risingrising
much much
while
less.
over theInequality
less.
wholeInequality
in Sub-Saharan
period in Sub-Saharan
2005-2018. Africa
These has
Africa has
variables in in t
in Southeast Asia. In the MENA region, inequality
that risen risen
risen arethat
GDP per capita, social support, healthy liferegion,
rose from 2009 toon in
the
on steep
2013, the steep
while post-2010
beingpost-2010
Southeast
pathsince.
stable similar
path
on similar
to tointhat
Southeast
in Southeast
Asia.
the In
Asia.
the MENA
In the MENA
region, stee A
expectancy, freedom, generosity, and absence of
ing inequality
inequality
inequality
stable roserose
fromfrom
20092009to 2013,
to 2013,
whilewhile
beingbeing
stablestable
since.
since.since. rose
corruption. Note that we do not construct our
happiness measure in each country using these
six factors – the scores are instead based on
Ranking Ranking
Ranking of Happiness
of Happiness by Country
by Country of H
individuals’ own assessments of their lives, as
vering
Now
Now
Now
we turn
we turn
to consider
to consider
lifethe
evaluations
life evaluations
we covering
covering
the 2016-2018
the
2016-2
2016-2018
turn
period,period,
and to present
and to present to
indicated by the Cantril ladder. Rather, we use The results in the first column of Table 2.1 explain
the six variables to explain the variation of national average life evaluations in terms of six key
happiness across countries. We shall also show how variables: GDP per capita, social support, healthy
measures of experienced well-being, especially life expectancy, freedom to make life choices,
positive affect, supplement life circumstances in generosity, and freedom from corruption.12 Taken
explaining higher life evaluations. together, these six variables explain almost
three-quarters of the variation in national annual
In Table 2.1 we present our latest modeling of
average ladder scores among countries, using
national average life evaluations and measures of
data from the years 2005 to 2018. The model’s
positive and negative affect (emotion) by country
predictive power is little changed if the year
and year.10 For ease of comparison, the table has
fixed effects in the model are removed, falling
the same basic structure as Table 2.1 in several 20
from 0.740 to 0.735 in terms of the adjusted
previous editions of the World Happiness Report.
R-squared.
The major difference comes from the inclusion of 21
data for 2018, and the resulting changes to the The second and third columns of Table 2.1 use
estimated equation are very slight.11 There are the same six variables to estimate equations
four equations in Table 2.1. The first equation for national averages of positive and negative
provides the basis for constructing the sub-bars affect, where both are based on answers
shown in Figure 2.7. about yesterday’s emotional experiences
Table 2.1: Regressions to Explain Average Happiness across Countries (Pooled OLS)
Dependent Variable
Cantril Ladder Positive Affect Negative Affect Cantril Ladder
Independent Variable (0-10) (0-1) (0-1) (0-10)
Log GDP per capita 0.318 -.011 0.008 0.338

(0.066)*** (0.01) (0.008) (0.065)***
Social support 2.422 0.253 -.313 1.977
(0.381)*** (0.05)*** (0.051)*** (0.397)***
Healthy life expectancy at birth 0.033 0.001 0.002 0.03
(0.01)*** (0.001) (0.001) (0.01)***
Freedom to make life choices 1.164 0.352 -.072 0.461
(0.3)*** (0.04)*** (0.041)* (0.287)
Generosity 0.635 0.137 0.008 0.351
(0.277)** (0.03)*** (0.028) (0.279)
Perceptions of corruption -.540 0.025 0.094 -.612
(0.294)* (0.027) (0.024)*** (0.287)**
Positive affect 2.063
(0.384)***
Negative affect 0.242
(0.429)
Year fixed effects Included Included Included Included
Number of countries 157 157 157 157
Number of obs. 1,516 1,513 1,515 1,512
Adjusted R-squared 0.74 0.476 0.27 0.76
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 2018. See Technical Box 1 for detailed information about each
of the predictors. Coefficients are reported with robust standard errors clustered by country in parentheses. ***, **,
and * indicate significance at the 1, 5 and 10 percent levels respectively.
(see Technical Box 1 for how the affect measures Our country rankings in Figure 2.7 show life
are constructed). In general, the emotional evaluations (the average answer to the Cantril
measures, and especially negative emotions, are ladder question, asking people to evaluate the
differently, and much less fully, explained by the quality of their current lives on a scale of 0 to 10)
six variables than are life evaluations. Per-capita for each country, averaged over the years
income and healthy life expectancy have significant 2016-2018. Not every country has surveys in
effects on life evaluations, but not, in these every year; the total sample sizes are reported
national average data, on either positive or in the statistical appendix, and are reflected in
negative affect. The situation changes when we Figure 2.7 by the horizontal lines showing the 95%
consider social variables. Bearing in mind that confidence intervals. The confidence intervals are
positive and negative affect are measured on a tighter for countries with larger samples. To
0 to 1 scale, while life evaluations are on a 0 to 10 increase the number of countries ranked, we also
scale, social support can be seen to have similar include three countries that did have surveys in
proportionate effects on positive and negative 2015 but have not had one since.15
emotions as on life evaluations. Freedom and
The overall length of each country bar represents
generosity have even larger influences on positive
the average ladder score, which is also shown in
affect than on the ladder. Negative affect is
numerals. The rankings in Figure 2.7 depend only
significantly reduced by social support, freedom,
on the average Cantril ladder scores reported by
and absence of corruption.
the respondents, and not on the values of the six
In the fourth column we re-estimate the life variables that we use to help account for the
evaluation equation from column 1, adding both large differences we find.
positive and negative affect to partially implement
Each of these bars is divided into seven segments,
the Aristotelian presumption that sustained
showing our research efforts to find possible
positive emotions are important supports for a
sources for the ladder levels. The first six sub-bars
good life.13 The most striking feature is the extent to
show how much each of the six key variables is
which the results buttress a finding in psychology
calculated to contribute to that country’s ladder
that the existence of positive emotions matters
score, relative to that in a hypothetical country
much more than the absence of negative ones.14
called Dystopia, so named because it has values
Positive affect has a large and highly significant
equal to the world’s lowest national averages for
impact in the final equation of Table 2.1, while
2016-2018 for each of the six key variables used
negative affect has none.
in Table 2.1. We use Dystopia as a benchmark
As for the coefficients on the other variables in against which to compare contributions from
the final equation, the changes are material only each of the six factors. The choice of Dystopia as
on those variables – especially freedom and a benchmark permits every real country to have
generosity – that have the largest impacts on a positive (or at least zero) contribution from
positive affect. Thus we infer that positive each of the six factors. We calculate, based on
emotions play a strong role in support of life the estimates in the first column of Table 2.1, that
evaluations, and that much of the impact of Dystopia had a 2016-2018 ladder score equal to
freedom and generosity on life evaluations is 1.88 on the 0 to 10 scale. The final sub-bar is the
channeled through their influence on positive sum of two components: the calculated average
emotions. That is, freedom and generosity have 2016-2018 life evaluation in Dystopia (=1.88) and
large impacts on positive affect, which in turn each country’s own prediction error, which
has a major impact on life evaluations. The measures the extent to which life evaluations are
Gallup World Poll does not have a widely higher or lower than predicted by our equation in
available measure of life purpose to test the first column of Table 2.1. These residuals are
whether it too would play a strong role in as likely to be negative as positive.16
support of high life evaluations. However, data
It might help to show in more detail how we
from large samples of UK do suggest that life
calculate each factor’s contribution to average
purpose plays a strongly supportive role,
life evaluations. Taking the example of healthy life
independent of the roles of life circumstances
expectancy, the sub-bar in the case of Tanzania
and positive emotions.
is equal to the number of years by which healthy
Technical Box 1: Detailed information about each of the predictors in Table 2.1
1. GDP per capita is in terms of Purchasing 5. G

enerosity is the residual of regressing
Power Parity (PPP) adjusted to constant the national average of GWP responses
2011 international dollars, taken from the to the question “Have you donated
World Development Indicators (WDI) money to a charity in the past month?”
released by the World Bank on November on GDP per capita.
14, 2018. See Statistical Appendix 1 for
6. P
erceptions of corruption are the average
more details. GDP data for 2018 are not 22
of binary answers to two GWP questions:
yet available, so we extend the GDP time
“Is corruption widespread throughout the 23
series from 2017 to 2018 using country-
government or not?” and “Is corruption
specific forecasts of real GDP growth
widespread within businesses or not?”
from the OECD Economic Outlook No.
Where data for government corruption
104 (Edition November 2018) and the
are missing, the perception of business
World Bank’s Global Economic Prospects
corruption is used as the overall corrup-
(Last Updated: 06/07/2018), after adjust-
tion-perception measure.
ment for population growth. The equation
uses the natural log of GDP per capita, as 7. Positive affect is defined as the average
this form fits the data significantly better of previous-day affect measures for
than GDP per capita. happiness, laughter, and enjoyment for
GWP waves 3-7 (years 2008 to 2012, and
2. T
he time series of healthy life expectancy
some in 2013). It is defined as the average
at birth are constructed based on data
of laughter and enjoyment for other
from the World Health Organization
waves where the happiness question was
(WHO) Global Health Observatory data
not asked. The general form for the
repository, with data available for 2005,
affect questions is: Did you experience
2010, 2015, and 2016. To match this
the following feelings during a lot of the
report’s sample period, interpolation and
day yesterday? See pp. 1-2 of Statistical
extrapolation are used. See Statistical
Appendix 1 for more details.
Appendix 1 for more details.
8. N
egative affect is defined as the average
3. Social support is the national average
of previous-day affect measures for
of the binary responses (either 0 or 1)
worry, sadness, and anger for all waves.
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 whenever you need them,
or not?”
4. F
reedom to make life choices is the
national average of binary responses to
the GWP question “Are you satisfied or
dissatisfied with your freedom to choose
what you do with your life?”
life expectancy in Tanzania exceeds the world’s life circumstances tend to average out at the
lowest value, multiplied by the Table 2.1 coefficient national level.
for the influence of healthy life expectancy on
To provide more assurance that our results are
life evaluations. The width of these different
not seriously biased because we are using the
sub-bars then shows, country-by-country, how
same respondents to report life evaluations, social
much each of the six variables is estimated to
support, freedom, generosity, and corruption,
contribute to explaining the international ladder
we tested the robustness of our procedure
differences. These calculations are illustrative
(see Table 10 of Statistical Appendix 1 of World
rather than conclusive, for several reasons. First,
Happiness Report 2018 for more detail) by
the selection of candidate variables is restricted
splitting each country’s respondents randomly
by what is available for all these countries.
into two groups, and using the average values for
Traditional variables like GDP per capita and
one group for social support, freedom, generosity,
healthy life expectancy are widely available. But
and absence of corruption in the equations to
measures of the quality of the social context,
explain average life evaluations in the other half
which have been shown in experiments and
of the sample. The coefficients on each of the
national surveys to have strong links to life
four variables fall, just as we would expect. But
evaluations and emotions, have not been
the changes are reassuringly small (ranging from
sufficiently surveyed in the Gallup or other
1% to 5%) and are far from being statistically
global polls, or otherwise measured in statistics
significant.19
available for all countries. Even with this limited
choice, we find that four variables covering The seventh and final segment is the sum of
different aspects of the social and institutional two components. The first component is a fixed
context – having someone to count on, generosity, number representing our calculation of the
freedom to make life choices and absence of 2016-2018 ladder score for Dystopia (=1.88).
corruption – are together responsible for more The second component is the average 2016-2018
than half of the average difference between each residual for each country. The sum of these two
country’s predicted ladder score and that in components comprises the right-hand sub-bar
Dystopia in the 2016-2018 period. As shown in for each country; it varies from one country to
Statistical Appendix 1, the average country has a the next because some countries have life
2016-2018 ladder score that is 3.53 points above evaluations above their predicted values, and
the Dystopia ladder score of 1.88. Of the 3.53 others lower. The residual simply represents that
points, the largest single part (34%) comes part of the national average ladder score that is
from social support, followed by GDP per capita not explained by our model; with the residual
(26%) and healthy life expectancy (21%), and included, the sum of all the sub-bars adds up to
then freedom (11%), generosity (5%), and the actual average life evaluations on which the
corruption (3%).17 rankings are based.
Our limited choice means that the variables we What do the latest data show for the 2016-2018
use may be taking credit properly due to other country rankings? Two features carry over from
better variables, or to unmeasured factors. There previous editions of the World Happiness Report.
are also likely to be vicious or virtuous circles, First, there is still a lot of year-to-year consistency
with two-way linkages among the variables. For in the way people rate their lives in different
example, there is much evidence that those who countries, and of course we do our ranking on a
have happier lives are likely to live longer, be three-year average, so that there is information
more trusting, be more cooperative, and be carried forward from one year to the next. But
generally better able to meet life’s demands.18 there are nonetheless interesting changes. The
This will feed back to improve health, GDP, annual data for Finland have continued their
generosity, corruption, and sense of freedom. modest but steady upward trend since 2014,
Finally, some of the variables are derived from so that dropping 2015 and adding 2018 boosts
the same respondents as the life evaluations and the average score, thereby putting Finland
hence possibly determined by common factors. significantly ahead of other countries in the top
This risk is less using national averages, because ten. Denmark and Norway have also increased
individual differences in personality and many their average scores, but Denmark by more than
Norway, so Denmark is now in second place and scores, and hence in country rankings, between
Norway third. There are no 2018 survey results 2005-2008 and 2016-2018, as will be shown in
available for Iceland, and their score and ranking more detail in Figure 2.8.
remain the same, in 4th place. The Netherlands
When looking at average ladder scores, it is also
have slipped into 5th place, dropping Switzerland
important to note the horizontal whisker lines at
to 6th.The next three places contain the same
the right-hand end of the main bar for each
three countries as last year, with Sweden’s
country. These lines denote the 95% confidence
increasing scores raising it to 7th, with New
regions for the estimates, so that countries with
Zealand remaining 8th and Canada now in 9th.
overlapping error bars have scores that do not
The final position in the top ten goes to Austria,
significantly differ from each other. The scores
rising from 12th to 10th, with Australia dropping
are based on the resident populations in each
to 11th, followed by Costa Rica in 12th, and Israel
country, rather than their citizenship or place of
in 13th.There are further changes in the rest of
birth. In World Happiness Report 2018 we split
the top 20, with Luxembourg rising to 14th and
the responses between the locally and foreign-
the United Kingdom to 15th, Ireland and Germany
born populations in each country, and found the
in 16th and 17th, and Belgium and the United
happiness rankings to be essentially the same for
States in 18th and 19th. The Czech Republic
the two groups, although with some footprint
rounds out the top 20 by switching positions
effect after migration, and some tendency for
with the United Arab Emirates. Both countries
migrants to move to happier countries, so that
posted rising averages, with the Czech score
among 20 happiest countries in that report, the
rising more. Throughout the top 20 positions,
average happiness for the locally born was about
and indeed at most places in the rankings, even
0.2 points higher than for the foreign-born.20
the three-year average scores are close enough
to one another that significant differences are Average life evaluations in the top 10 countries
found only between country pairs that are are more than twice as high as in the bottom 10.
several positions apart in the rankings. This If we use the first equation of Table 2.1 to look
can be seen by inspecting the whisker lines for possible reasons for these very different
showing the 95% confidence intervals for the life evaluations, it suggests that of the 4.16 points
average scores. difference, 3.06 points can be traced to differences
in the six key factors: 0.99 points from the GDP
There remains a large gap between the top and
per capita gap, 0.88 due to differences in social
bottom countries. The top ten countries are less
support, 0.59 to differences in healthy life
tightly grouped than last year. The national life
expectancy, 0.35 to differences in freedom,
evaluation scores now have a gap of 0.28 between
0.20 to differences in corruption perceptions,
the 1st and 5th position, and another 0.24 between
and 0.06 to differences in generosity.21 Income
5th and 10th positions, a more spread-out
differences are the single largest contributing
situation than last year. Thus there is now a gap
factor, at one-third of the total, because, of the
of about 0.5 points between the first and 10th
six factors, income is by far the most unequally
positions. There is a bigger range of scores
distributed among countries. GDP per capita is
covered by the bottom 10 countries. Within this
22 times higher in the top 10 than in the bottom
group, average scores differ by almost
10 countries.22
three-quarters of a point, more than one-fifth of
the average national score in the group. Tanzania, Overall, the model explains average life
Rwanda and Botswana still have anomalous evaluation levels quite well within regions,
scores, in the sense that their predicted values, among regions, and for the world as a whole.23
based on their performance on the six key On average, the countries of Latin America still
variables, would suggest they would rank much have mean life evaluations that are higher (by
higher than shown by the survey answers. about 0.6 on the 0 to 10 scale) than predicted
by the model. This difference has been attributed
Despite the general consistency among the top
to a variety of factors, including especially some
country scores, there have been many significant
unique features of family and social life in Latin
changes in the rest of the countries. Looking at
American countries. To help explain what is
changes over the longer term, many countries
special about social life in Latin America,
have exhibited substantial changes in average
Chapter 6 of World Happiness Report 2018 by grouped in the same global regions used
Mariano Rojas presented a range of new data elsewhere in the report. Within leagues, countries
and results showing how the social structure are ordered by their 2016-2018 ladder scores.
supports Latin American happiness beyond
what is captured by the variables available in
the Gallup World Poll. In partial contrast, the
countries of East Asia have average life
evaluations below those predicted by the model,
a finding that has been thought to reflect, at
least in part, cultural differences in response
style.24 It is reassuring that our findings about
the relative importance of the six factors are 28
generally unaffected by whether or not we make

29
explicit allowance for these regional differences.25
Our main country rankings are based on the

average answers to the Cantril ladder life
evaluation question in the Gallup World Poll. The
other two happiness measures, for positive and
negative affect, are themselves of independent
importance and interest, as well as being,
especially in the case of positive affect,
contributors to overall life evaluations. Measures
of positive affect also play important roles in
other chapters of this report, in large part
because most lab experiments, being of
relatively small size and duration, can be
expected to affect current emotions but not life
evaluations, which tend to be more stable in
response to small or temporary disturbances.
The various attempts to use big data to measure
happiness using word analysis of Twitter feeds,
or other similar sources, are likely to be
capturing mood changes rather than overall life
evaluations. In this report, for the first time since
2012, we are presenting, in Table 2.2, rankings for
all three of the measures of subjective well-being
that we track: the Cantril ladder (and its standard
deviation, which provides a measure of happiness
inequality), positive affect and negative affect.
We also show country rankings for the six
variables we use in Table 2.1 to explain our
measures of subjective well-being.26 The same
data are also shown in graphical form, on a
variable by variable basis, in Figures 16 to 39
of Statistical Appendix 1. The numbers shown
reflect each country’s global rank for the variable
in question, with the number of countries ranked
depending on the availability of data. The league
tables are divided into a premier league (the
OECD, whose 36 member countries include 19 of
the top 20 countries) and a number of regional
leagues comprising the remaining countries
Changes in National Happiness and reductions exceeding about 0.5 points, seven
Its Main Supports are in the Middle East and North Africa, six in
Sub-Saharan Africa, three in Western Europe,
We now turn to our country-by-country ranking with the remaining significant losers being
of changes in life evaluations. In the two previous Venezuela, India, Malaysia and Ukraine.
reports, we concentrated on looking at recent
changes in life evaluations. This year we take These changes are very large, especially for the
advantage of the ever-growing length of the 10 most affected gainers and losers. For each of
Gallup sample to compare life evaluations the 10 top gainers, the average life evaluation
over a longer span, averaging ten years, from gains were more than would be expected from a
2005-2008 to 2016-2018. In Figure 2.8 we show tenfold increase of per capita incomes. For each
the changes in happiness levels for all 132 countries of the 10 countries with the biggest drops in
that have sufficient numbers of observations for average life evaluations, the losses were more
both 2005-2008 and 2016-2018. than twice as large as would be expected from
a halving of GDP per capita.
Of the 132 countries with data for 2005-2008
and 2016-2018, 106 had significant changes. 64 On the gaining side of the ledger, the inclusion of
were significant increases, ranging from 0.097 to four transition countries among the top 10
1.39 points on the 0 to 10 scale. There were also gainers reflects the rising average life evaluations
42 significant decreases, ranging from -0.179 to for the transition countries taken as a group. The
–1.944 points, while the remaining 26 countries appearance of Sub-Saharan African countries
revealed no significant trend from 2005-2008 to among the biggest gainers and the biggest
2016-2018. As shown in Table 32 in Statistical losers reflects the variety and volatility of
Appendix 1, the significant gains and losses are experiences among the Sub-Saharan countries
very unevenly distributed across the world, and for which changes are shown in Figure 2.8, and
sometimes also within continents. In Central and whose experiences were analyzed in more detail
Eastern Europe, there were 15 significant gains in Chapter 4 of World Happiness Report 2017.
against only one significant decline, while in Benin, the largest gainer since 2005-2008, by
Western Europe there were eight significant almost 1.4 points, ranked 4th from last in the
losses compared to four significant gains. The first World Happiness Report and has since risen
Commonwealth of Independent States was a 50 places in the rankings.
significant net gainer, with eight gains against The 10 countries with the largest declines in
two losses. In Latin America and the Caribbean average life evaluations typically suffered some
and in East Asia, significant gains outnumbered combination of economic, political, and social
significant losses by more than a two to one stresses. The five largest drops since 2005-2008
margin. The Middle East and North Africa was were in Yemen, India, Syria, Botswana and
net negative, with six losses against three gains. Venezuela, with drops over one point in each
In the North American and Australasian region, case, the largest fall being almost two points in
the four countries had two significant declines Venezuela. Among the countries most affected
and no significant gains. The 28 Sub-Saharan by the 2008 banking crisis, Greece is the only
African countries showed a real spread of one remaining among the 10 largest happiness
experiences, with 13 significant gainers and 10 losers, although Spain and Italy remain among
significant losers. In South and Southeast Asia, the 20 largest.
most countries had significant changes, with a
Figure 42 and Table 31 in Statistical Appendix 1
fairly even balance between gainers and losers.
show the population-weighted actual and
Among the 20 top gainers, all of which showed predicted changes in happiness for the 10
average ladder scores increasing by more than regions of the world from 2005-2008 to
0.7 points, 10 are in the Commonwealth of 2016-2018. The correlation between the actual
Independent States or Central and Eastern and predicted changes is only 0.14, and with
Europe, five are in Sub-Saharan Africa, and actual changes being less favorable than predicted.
three in Latin America. The other two are Only in Central and Eastern Europe, where life
Pakistan and the Philippines. Among the evaluations were up by 0.6 points on the 0 to 10
20 largest losers, all of which show ladder scale, was there an actual increase exceeding
what was predicted. South Asia had the largest shown forcefully in World Happiness Report
drop in actual life evaluations (more than 0.8 2018, which presented happiness rankings for
points on the 0 to 10 scale) while it was predicted immigrants and the locally born, and found them
to have a substantial increase. Since these to be almost exactly the same (a correlation of
regional averages are weighted by national +0.96 for the 117 countries with a sufficient
populations, the South Asian total is heavily number of immigrants in their sampled
influenced by the Indian decline of more than 1.1 populations). This was the case even for
points. Sub-Saharan Africa was predicted to have migrants coming from source countries with
a substantial gain, while the actual increase was life evaluations less than half as high as in the
much smaller. Latin America was predicted to destination country.
have a small gain, while it shows a popula-
The evidence from the happiness of immigrants
tion-weighted actual drop of the same size. The
and the locally born suggests strongly that the
MENA region was predicted to be a gainer, and
large international differences in average national
instead lost 0.52 points. The countries of Western
happiness documented in this report depend
Europe were predicted to have no change, but
primarily on the circumstances of life in each
instead experienced a small reduction. For the
country. These differences in turn invite explanation
remaining regions, the predicted and actual
by factors that differ among nations, including
changes were in the same direction, with the
especially institutions that are national in scope,
substantial reductions in the United States (the
among which governments are perhaps the most
largest country in the NANZ group) being larger
prominent examples.
than predicted. As Figure 42 and Table 31 show
in Statistical Appendix 1, changes in the six It is natural, as public and policy attention starts to
factors are not very successful in capturing the shift from GDP to broader measures of progress,
evolving patterns of life over what have been and especially to how people value their lives,
tumultuous times for many countries. Nine of the that there should be growing policy interest in
ten regions were predicted to have 2016-2018 life knowing how government institutions and
evaluations higher than in 2005-2008, but only actions influence happiness, and in whatever
half of them did so. In general, the ranking of changes in policies might enable citizens to lead
regional predicted changes matched the ranking happier lives.
of the actual changes, despite typical experience
being less favorable than predicted. The notable
exception is South Asia, which experienced the What is Good Government?
largest drop, contrary to predictions.
At the most basic level, good government
On a country-by-country basis, the actual changes establishes and maintains an institutional
from 2005-2008 to 2016-2018 are on average framework that enables people to live better
much better predicted than on a regional basis, lives. Similarly, good public services are those
with a correlation of 0.50, as shown in Figure 41 that improve lives while using fewer scarce
in Statistical Appendix 1. This difference can be resources. How can the excellence of govern-
traced to the great variety of experiences within ment be measured, and how can its effects on
regions, many of which were predicted reasonably happiness be determined? There are two main
well on a national basis, and by the presence of possibilities for assessment, one very specific
some very large countries with substantial and the other at the aggregate level. The more
prediction errors, India being the largest example. specific approach is adopted in the Global
Happiness and Well-being Policy Reports, while
here we shall take a more aggregate approach
Changes in Governance using the national happiness data that lie at the
core of the World Happiness Reports.
Government institutions and policies set the
stages on which lives are lived. These stages differ Created in response to growing interest in the
largely from country to country, and are among policy relevance of happiness, the Global Happiness
primary factors influencing how highly people and Well-being Policy Reports aim to find and
rate the quality of their lives. The importance of evaluate best-practice examples from around the
national institutions and living conditions was world on how government policies in specific
areas could be redesigned to support happier this sort of research. We consider here some of
lives. The just-released Global Happiness and the effects of government structure and behavior
Well-being Policy Report 2019,27 for example, on average national happiness, while Chapter 3
contains surveys of happiness-oriented policy considers how happiness affects voting behavior.
interventions in specific areas of public policy –
Our own analysis in Table 2.1 provides one
in particular education, health, work and cities
example of the effects of government via its
– as well as on topics of cross-cutting importance,
estimate of the links between corruption and life
such as personal happiness28 and the metrics and
satisfaction, holding constant some other key
policy frameworks29 needed to support policies
variables, including income, health, social
for well-being. These policy surveys show that
support, a sense of freedom and generosity, all
what counts as good governance is specific to
of which themselves are likely to be affected by 38
each policy area. Within each ministry or subject
the quality of government. Unpacking these
area there are specific targets that are the
channels convincingly is not possible using the 39
primary focus of attention, including mainly
aggregate data available, since there is too much
medical and cost outcomes in health care,30
in play to establish strong evidence of causality,
academic achievement and completion in
and many of the system features held to be of
education,31 productivity and job satisfaction in
primary importance, for example the rule of law,
the workplace,32 reduced crime and incarceration
tend to take long to establish, thereby reducing
rates in justice, and a range of specific indicators
the amount of evidence available.
of the quality of city life.33 The happiness lens is
then used to find those policies that achieve their Hence any conclusions reached are likely to be
traditional objectives in the most happiness- suggestive at best, and have also been found to
supporting ways. This kind of specific focus is be more evident in some countries and times
probably the most effective way to move from a than in others. For example, a number of studies
general interest in using happiness as a policy have divided the World Bank’s37 six main
objective to the development of cost-effective indicators of governmental quality into two
ways of delivering happiness. One major common groups, with the four indicators for effectiveness,
element in the chapters of Global Happiness rule of law, quality of regulation, and control of
and Well-being Policy Report 2019 is the use of corruption combined to form an index of the
results from happiness research to establish the quality of delivery, and the two indicators for
relative importance of a variety of outcomes long voice and accountability and for political stability
considered important but not readily comparable. and absence of violence combined to form an
As advocated by Chapter 634 in World Happiness index of the democratic quality of government.
Report 2013, developed in more detail in a recent
Previous studies comparing these two indexes
paper35 for the UK Treasury, and exemplified by
as predictors of life evaluations have found that
the happiness-based policy evaluation tool in
quality of delivery is more important than the
Dubai, and in the health chapter36 of Global
democracy variable, both in studies across
Happiness and Well-being Policy Report 2019,
countries38 and in ones that include country-fixed
this involves expanding traditional methods for
effects, so that the estimated effects are based
estimating the cost-effectiveness of policies to
on changes in governance quality within each
make happier lives the objective. Seen from this
country.39 These latter results are more convincing,
perspective, good governance would be defined
since they are uninfluenced by cross-country
in terms of the methods used and results
differences in other variables, and have the
obtained, both for traditional policy objectives
capacity to show whether significant changes in
and the happiness of all participants.
the quality of government can happen within a
There is another way of assessing different policy-relevant time horizon. These studies made
government structures and policies. This is done use of data from the World Values Survey and
at a more aggregate level by using a number from the Gallup World Poll, but were based on
of national-level indicators of the quality of shorter sample periods. For this chapter we
governance to see how well they correlate with replicated earlier analysis based on the GWP
levels and changes in national average life data for 2005-2012 but now using the longest
evaluations. There are now many examples of sample with available data for life evaluations
and for the indicators of government quality, effect disappears in the new longer sample,
covering 2005-2017. The results are shown in where we find that changes in the quality of
Table 10 in Statistical Appendix 2. The core delivery have equally large and significant effects
results continue to show that delivery quality on life evaluations, and changes in democratic
has a significant positive effect on average life quality have no significant effects, whatever the
evaluations with or without accounting for the average state of delivery quality.42
effects flowing through the higher levels of
Tables 12 to 15 in Statistical Appendix 2 test
GDP per capita made possible by government
whether changes in a variety of other measures
regulations and services that are more efficient,
of governmental quality contribute to changes in
more configured to match the rule of law, and
life evaluations. None show significant effects
less subject to corruption. The estimated
with one notable exception. Changes in the
magnitude of the more convincing results, which
Gallup World Poll’s measure of confidence in
are the ones based on within-country changes in
government do contribute significantly to life
governance quality, is substantial. For example, a
evaluations, as shown in Table 13 of Statistical
previous study found that “the ten most-improved
Appendix 2. To some extent, this variable might
countries, in terms of delivery quality changes
be thought to reflect a measure of satisfaction
between 2005 and 2012, when compared to the
with a particular life domain, much as was shown
ten countries with most worsened delivery
in Figure 1.1 for Mexico in Chapter 1.
quality, are estimated to have higher average life
evaluations by one tenth of the overall life Tables 16 to 18 of Statistical Appendix 2 look for
evaluation gap between the world’s most and linkages between average life evaluations and a
least happy ten countries.”40 In other words, the number of government characteristics including
estimated effect of the divergence in governance different forms of democratic institutions, social
quality on life evaluations was about 0.4 points safety net coverage, and percent of GDP devoted
on the 0 to 10 scale. We have been able to to education, healthcare and military spending.43
confirm that previous result with data now The only characteristics that contribute beyond
covering twice as long a time period, as shown what is explained by the six variables of Table 2.1
in Table 22 in Statistical Appendix 2. and regional fixed effects are the shares of GDP
devoted to healthcare and military spending, the
To extend our analysis into other aspects of
former having a positive effect and the latter a
governance, we have assembled data to match
negative one.44
our mix of country-years for several variables
that have either been used as measures of the It is noteworthy that many countries with low
quality of governance, or can been seen to average life evaluations, and with life evaluations
reflect some aspects of governmental quality. much lower than would be predicted by the
One question of perennial research and policy standard results in Table 2.1, have been subject to
interest is whether people are happier living in internal and external conflicts. Such conflicts can
political democracies. Our earlier research based in part be seen as evidence of bad governance,
on World Values Survey data and shorter samples and have no doubt contributed to bad governance
of Gallup World Poll data found that delivery elsewhere. In any event, they are almost surely
quality was always more important than the likely to lead to low life evaluations.45 For example,
measure of democratic quality, whether or not freedom from violence is part of one of the
the analysis included country fixed effects, which World Bank’s six indicators for the quality of
help to make the results more convincing. This is governance, and several of the countries among
still borne out in our doubled sample length for those ranked as least happy in Figure 2.7 are or
the Gallup World Poll (Table 10, Appendix 2). We have been subject to fatal political violence. We
also found in earlier research that if the sample have assembled data for several measures of
was split between countries with higher and internal and international conflict, and have
lower governmental effectiveness, that increases found evidence that conflict is correlated with
in the extent of democracy had positive life lower life evaluations, sometimes beyond what is
satisfaction effects in those countries with already captured by the variables for income,
effective governments, but not in countries with health, freedom, social support, generosity and
less effective governments.41 But this interaction corruption. The Uppsala data for death rates
from armed conflicts, non-state conflicts and 2005-2008 to 2016-2018 changes in life
one-sided violence are negatively correlated with evaluations, positive and negative affect, and the
life evaluations, but also with GDP per capita, the key variables supporting life evaluations. Finally,
World Bank’s democracy variables, and both we considered different ways in which the
freedom and social support. These correlations nature and quality of government policies and
are almost unchanged if put on a within-country institutions can influence happiness.
change basis, as can been seen by comparing
At a global level, population-weighted life
Tables 2 and 3 in Statistical Appendix 2. The
evaluations fell sharply during the financial crisis,
estimated impact of conflict deaths on average
recovered completely by 2011, and fell fairly
life evaluations is especially great in the 14
steadily since to a 2018 value about the same
countries where conflict deaths have in one or
level as its post-crisis low. This pattern of falling 40
more years been above the 90th percentile of
global life evaluations since 2011 was driven
the distribution of positive death rates by year
mainly by what was happening in the five 41
from 2005 to 2017.46 But even here they add little
countries with the largest populations, and
additional explanatory power once allowance is
especially India, which has had a post-2011 drop
made for all the other variables in Table 2.1.
of almost a full point on the 0 to 10 scale.
Somewhat stronger results are obtained by using Excluding the five largest countries removes
the Global Peace Index assessing 163 countries in the decline, while an unweighted average of the
three domains: the level of societal safety and country scores shows a significant rise since
security, the extent of ongoing domestic and 2016. Positive emotions show no significant
international conflict, and the degree of militari- trends by either weighted or unweighted
sation. The index (which is defined as if it were a measures. Negative emotions show the most
conflict variable, so that a more peaceful country dramatic global trends, rising significantly by
has a lower value) is negatively correlated with both global measures. Global inequality of
average life evaluations in both levels and changes well-being has been fairly constant between
from 2008 to 2016-2018.47 The effect of within- countries while rising within countries.
country changes in the peace index remains
These global movements mask a greater variety
significant even when changes in GDP and the
of experiences among and within global regions.
rest of the six key variables are included, with a
There continues to be convergence of life
change of 0.5 in the peace index (about 1 standard
evaluations among the three main regions of
deviation) estimated to alter average life
Europe. In Asia, divergence among the regions is
evaluations by 0.15 points on the 0 to 10 scale,
more evident. All three parts of Asia had roughly
a value equivalent to a change of more than
comparable life evaluations in the 2006-2010
15% in per capita GDP.48
period, but since then life evaluations have
generally risen in East and Southeast Asia and
fallen in South Asia, with a gap building to more
Conclusions than 1 point on the 0 to 10 scale by 2018. Since
This chapter has had a special focus on how 2013, life evaluations have risen by 0.4 points in
several measures of happiness, and of its Sub-Saharan Africa and fallen by 0.4 points in
contributing factors, have changed over the the Middle East and North Africa, finishing in
2005 to 2018 period covered by the Gallup 2018 at roughly equal levels. In Latin America, life
World Poll. We started by tracing the trajectories evaluations rose by half a point to 2013, and have
of happiness, and its distribution, primarily based fallen slightly more than that since, while in the
on annual population-weighted averages for the North America plus Australia and New Zealand
world as a whole and for its ten constituent group, with population dominated by the United
regions. This was followed by our latest ranking States, life evaluations have fallen by roughly
of countries according to their average life 0.3 points from the beginning to the end of
evaluations over the previous three years, the period.
accompanied this year by comparable rankings
What about well-being inequality? Since 2012,
for positive and negative affect, for six key
the mid-point of our data period, well-being
factors used to explain happiness, and for
inequality has fallen insignificantly in Western
happiness inequality. We then presented
Europe and Central and eastern Europe, while

increasing significantly in most other regions,
including especially South Asia, Southeast Asia,
Sub-Saharan Africa, the Middle East and North
Africa, and the CIS (with Russia dominating the
population total).
The rankings of country happiness are based this

year on the pooled results from Gallup World
Poll surveys from 2016-2018, and continue to
show both change and stability. As shown by
our league tables for happiness and its supports,
the top countries tend to have high values for
most of the key variables that have been found
to support well-being: income, healthy life
expectancy, social support, freedom, trust and
generosity, to such a degree that year to year
changes in the top rankings are to be expected.
With its continuing upward trend in average
scores, Finland consolidated its hold on first
place, ahead of an also-rising Denmark in
second place.
Then for each country, we showed that average

changes in life evaluations from the earliest years
of the Gallup World Poll (2005-2008) to the
three most recent years (2016-2018). Most
countries show significant changes, with slightly
more gainers than losers. The biggest gainer
was Benin, up 1.4 points and 50 places in the
rankings. The biggest life evaluation drops were
in Venezuela and Syria, both down by about
1.9 points.
We turned finally to consider the ways in which

the quality of government, and the structure of
government policies, influence happiness. The
effects were seen to be easier to trace in specific
policy areas, but also showed up in aggregate
measures of governmental quality, whether
based on citizen perceptions or the quality
indicators prepared by the World Bank. Among
these latter measures, the greatest impact still
appears to flow from the quality of policy delivery,
including the control of corruption. Finally,
making use of international data measuring
peace and conflict, countries able to reduce
conflict and achieve peace were estimated to
become happier places to live.

WHR19 - Ch2 VF PDF

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

WHR19 - Ch2 VF PDF

Uploaded by

Copyright:

Available Formats

Introduction The Evolution of World Happiness

In the first World Happiness Report we surveyed 2005-2018

.3 5.3 5.3 5.3 0.72 0.72 0.72

0.20 0.20 0.20  Non-population weighted

85 0.85 0.85

Figure 2.5: Dynamics of Inequality of Ladder (Standard Deviation)

Inequality of World Happiness

World Happiness Report 2019

Figure 2.6: Dynamics of Inequality of Ladder in 10 Regions

0 2.00 2.00 2.00 2.00 2.00

Asia Asia Africa and Middle East

0 2.00 2.00 2.00 2.00 2.00 Middle East

 East Asia  Middle East and North Africa

Log GDP per capita 0.318 -.011 0.008 0.338

1. GDP per capita is in terms of Purchasing 5. G

generally unaffected by whether or not we make

Our main country rankings are based on the

Europe and Central and eastern Europe, while

The rankings of country happiness are based this

Then for each country, we showed that average

We turned finally to consider the ways in which

You might also like

WHR19 - Ch2 VF PDF

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

WHR19 - Ch2 VF PDF

Uploaded by

Copyright:

Available Formats

Introduction The Evolution of World Happiness

In the first World Happiness Report we surveyed 2005-2018

.3 5.3 5.3 5.3 0.72 0.72 0.72

0.20 0.20 0.20  Non-population weighted

85 0.85 0.85

Figure 2.5: Dynamics of Inequality of Ladder (Standard Deviation)

Inequality of World Happiness

World Happiness Report 2019

Figure 2.6: Dynamics of Inequality of Ladder in 10 Regions

0 2.00 2.00 2.00 2.00 2.00

Asia Asia Africa and Middle East

0 2.00 2.00 2.00 2.00 2.00 Middle East

 East Asia  Middle East and North Africa

Log GDP per capita 0.318 -.011 0.008 0.338

1. GDP per capita is in terms of Purchasing 5. G

generally unaffected by whether or not we make

Our main country rankings are based on the

Europe and Central and eastern Europe, while

The rankings of country happiness are based this

Then for each country, we showed that average

We turned finally to consider the ways in which

You might also like

1. GDP per capita is in terms of Purchasing 5. G