Billionaires, millionaires, inequality, and happiness · Billionaires, millionaires, inequality,...

Expert Comment

Billionaires, millionaires, inequality, and happiness

Picture credit: Dilok/Bigstock.com

Copyright © 2019 by Dialogue of Civilizations Research Institute The right of Vladimir Popov to be identified as the author of this publication is hereby asserted. The views and opinions expressed in this publication are those of the original author(s) and do not necessarily represent or reflect the views and opinions of the Dialogue of Civilizations Research Institute, its co-founders, or its staff members. All rights reserved. No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the publisher, except in the case of brief quotations embodied in critical reviews and certain other noncommercial uses permitted by copyright law. For permission requests, please write to the publisher: Dialogue of Civilizations Research Institute gGmbH Französische Straße 23 10117 Berlin Germany +49 30 209677900 [email protected]

Dialogue of Civilizations Research Institute

1

Billionaires, millionaires, inequality, and happiness

By Vladimir Popov

This paper is the logical continuation of my two 2018 papers ‘Paradoxes of happiness. Why do people feel more

comfortable with high levels of inequality and high murder rates?’ DOC Expert Comment, 18 June 2018 and ‘Why do some

countries have more billionaires than others? Explaining variations in the billionaire-intensity of GDP’. DOC Expert

Comment, 24 July 2018. My thanks go to Tony Shorrocks who kindly provided me with the excel table of the GWR

database. I am also most grateful to Elena Sulimova for collecting the data and preparing the database for the analysis.

Abstract

The relationship between inequality and happiness is counterintuitive. This applies to both

inequality in income and wealth distribution overall and also inequality at the very top of the

wealth pyramid, as measured by billionaire intensity (the ratio of billionaire wealth to GDP).

First, billionaire intensity appears to be higher in countries with low, not high, levels of income

inequality. Second, happiness indices are higher in countries with high percentages of

billionaire and millionaire wealth as a proportion of GDP, but with low levels of income

inequality.

This paper uses databases from the Forbes billionaires list, the Global Wealth Report

(GWR), and the World Happiness Report, as well as from the World Database on

Happiness. Using these datasets, I examine the relationship between income inequality and

happiness for over 200 countries from 2000 to 2018.

It turns out that in relatively poor countries – below $20,000-$30,000 per capita

income – inequality raises happiness rather than lowers it, but inequality has a negative

impact on happiness in rich countries. A certain degree of inequality of wealth and income

distribution has a positive impact on happiness feelings, especially in countries with low

levels of income. Furthermore, wealth inequalities, and especially the degree of

concentration of wealth at the very top of the wealth pyramid, raise happiness self-

evaluations even when income inequalities lower it.

JEL: D31, D63, I31

Literature review, puzzles, and hypotheses

https://pages.nes.ru/vpopov/documents/Paradoxes%20of%20happiness.pdf

https://pages.nes.ru/vpopov/documents/Paradoxes%20of%20happiness.pdf

http://pages.nes.ru/vpopov/documents/Billionaires_Web-upload.pdf




2

There are some important paradoxes in the dynamics of happiness indices and the relative

levels of these indices for various countries and different populations groups. One puzzle,

the Easterlin paradox, is the non-increasing level of happiness in the US in spite of

constantly rising personal incomes (fig. 1).1

Figure 1: Average happiness score (left scale) and GDP per capita, dollars, (right

scale) in the US in 1972- 2016

Source: Sachs (2018).

Sachs (2018) argues that America’s subjective wellbeing is being systematically

undermined by three interrelated epidemics, notably obesity, substance abuse (especially

opioid addiction), and depression. But in other countries without as much obesity, drugs,

and depression, there is also a decline in happiness that goes hand in hand with rising real

incomes.

In India, the happiness index score fell from 5 to 4 over the 2006-18 period despite

strong growth of income in this period. In China, between 1990 and 2000, happiness

plummeted despite massive improvements in material living standards. Brockmann, Delhey,

Welzel, and Hao (2008) explain this by growing income inequality within China, i.e., in

relation to the average national income, the financial position of most Chinese people

deteriorated.

Similarly, in the US the recent increase in income inequalities could be responsible

for the decline in happiness: in 1980-2014, the post-tax incomes of the richest 10% rose by


3

113%; of the top 1%, by 194%; and of the top 0.001%, by 617% (Piketty, Saez, Zucman,

2016), whereas the US happiness index score over this period fell. However, the relationship

between inequality and happiness is also not straightforward and presents another puzzle.

Normally we assume that greater equality – ‘inclusive development that leaves no

one behind’ – is both morally just and desirable for the creation of happy societies. But there

is evidence that income and wealth inequalities are positively associated with happiness, as

measured by the happiness index, at least for a group of countries. There are some poorer

countries with high income inequalities – Bolivia, Honduras, Colombia, Ecuador, Costa Rica

and some other Latin American countries – and yet also with very high happiness index

scores (fig. 2). It may well be that a certain degree of inequality is necessary to keep alive a

kind of ‘American dream’: a future-oriented belief in getting rich and achieving success in

life.

Alesina, Di Tella, and MacCulloch (2001) showed that there is a large, negative, and

significant effect of inequality on happiness in Europe, but not in the US. It is also clear that

people have different perceptions of ‘correct’, ‘optimal’, or ‘just/fair’ degrees of inequality.

Alesina and La Ferrara (2001) found that individual support for redistribution is negatively

affected by social mobility. People who believe that American society offers equal

opportunities to all are more averse to redistribution in the face of increased mobility.

On the other hand, those who see the social rat race as a biased process do not see

social mobility as an alternative to redistributive policies. Alesina and Giuliano (2009)

presented evidence that individuals who believe other people try to take advantage of them

rather than being fair have a strong desire for redistribution; similarly, believing that luck is

more important than work as a driver of success is strongly associated with a taste for

redistribution.

Inequality at the very top does not seem to lead to pressure for redistribution. In

Victorian England, inequality within the elite was associated with more conditionality and

less generous welfare expenditure. Removing institutional advantages that benefited the

elite did not appear to reduce the effect of elite inequality, which suggests results cannot be

explained by a classic median voter model (Chapman, 2018).2 Therefore, an increase in

inequality at the very top may be self-perpetuating; there is no pressure to redistribute and

no mechanism to automatically ‘correct’ the inequality.


4

Figure 2: Happiness index (World Happiness Report and World Database on

Happiness), 0-to-10 scale; and GINI coefficient of income inequality in percentage

terms, 2000-18

Source: World Database on Happiness (2019); WDI.

There is also some evidence that happiness is positively correlated with murder rates,

especially when this goes hand in hand with inequalities (Popov, 2018a; 2018b). Inequalities

lead to higher murder rates, but this does not lead to a decline in happiness, at least up to

a certain point. Furthermore, happiness scores also seem almost independent of suicide

rates, which is often considered an objective indicator of happiness, in contrast to happiness

indices that measure self-perception through surveys where people measure their own

happiness on a 0-to-10 scale.

The relationship between self-reported happiness and objective indicators of

frustration and distress is somewhat counterintuitive. Murder rates are correlated positively

with happiness index scores (fig. 3), although this correlation is not statistically significant,

whereas the correlation between suicide rates and happiness is negative (fig. 4), as one

would expect but very weak: in regressions of happiness on murder rates, suicides rates,

and per capita income, R2 is less than 2% and the suicide rate is significant only in random

effects specification – it is apparent with the naked eye from a comparison of figure 4, which

presents cross-section data for the single year of 2018, and figure 5, which presents panel

data for the years 2000-18.3

ALBDZA

AGO

ARG

ARGARGARGARG

ARG

ARG

ARG

ARG

ARGARGARGARG

ARM

ARMARM

ARMARMARMARMARM

ARMARMARM

AUSAUSAUS

AUTAUTAUTAUTAUTAUTAUTAUTAUTAUT

AUTAUTAUT

BGDBGD

BLR

BLRBLRBLRBLR

BEL

BELBELBELBELBELBELBELBELBELBELBEL

BEL

BEN

BEN

BTN

BOL

BOL

BOLBOL

BOL

BOL

BOL

BOL

BOL

BOL

BOLBOLBOL

BIHBIHBIH

BRA

BRABRA

BRABRABRABRABRABRABRA

BGRBGRBGRBGRBGRBGRBGR

BGR

BGR

BFA

BDI

BDI

CMR CMR

CAN

CAN

TCD

CHL

CHL

CHL

CHLCHL

CHLCHL

CHNCHN

COLCOL

COL

COLCOLCOLCOL

COL

COL

COLCOLCOL

COM

COD COG

CRICRICRI

CRICRI

CRICRI

CRI

CRICRI

CRI

CRICRICRICRI

CIV

CIV

HRVHRVHRV

HRVHRVHRV

HRV

CYP

CYPCYPCYPCYPCYPCYPCYPCYP CYP

CYP

CYPCZE

CZECZECZECZECZECZECZE

CZECZECZECZE

DNKDNKDNKDNKDNK

DNKDNKDNKDNK

DNK

DNK

DNK

DJI

DOM

DOM

DOM

DOM

DOM

DOM

DOM

DOMDOMDOM

ECU

ECUECU

ECU

ECU

ECU

ECU

ECU

ECU

ECUECUECUECU

EGYEGY

EGYEGY

SLV

SLV

SLVSLV

SLV

SLV

SLV

SLV

SLV

SLV

SLV

SLVSLVSLV

EST

ESTESTESTESTEST

ESTESTESTESTEST

EST

EST

ETH ETH

FIN

FINFINFINFINFIN

FINFINFINFIN

FINFINFIN

FRAFRAFRAFRA FRA

FRA

FRA

FRAFRAFRAFRAFRAFRA

GAB

GEOGEO

GEO

GEOGEOGEOGEOGEOGEOGEOGEO

DEUDEUDEUDEUDEUDEU

DEUDEUDEU

GHA

GRCGRCGRCGRC

GRCGRC

GRC

GRCGRC

GRC

GRC

GRC

GRC

GTMGTM

GIN

GIN

HND

HND

HND

HND

HND

HND

HND

HND

HND

HND

HND

HNDHND

HUNHUNHUN

HUNHUN

HUNHUNHUNHUNHUN

HUN

HUN

ISL

IND

IDN IRNIRN

IRN

IRQ

IRQ

IRL

IRLIRLIRLIRLIRL IRLIRLIRLIRL

IRLIRLIRL ISR

ISRISR

ITAITAITAITAITA

ITAITA

ITAITA

ITA

ITA

ITAITA

JPNJOR

JOR

JOR

JORKAZKAZ

KAZ

KAZKAZKAZKAZKAZ

KAZKAZ

KEN

KORKOR

KORKOR

KOS

KOS

KOSKOS

KOS

KOSKOS

KOSKOS

KGZKGZKGZ

KGZKGZKGZKGZ

KGZKGZ KGZKGZ

LAO

LAOLVA

LVALVALVALVA

LVALVALVALVA

LVA

LVA

LVALBN

LBR

LBR

LTULTULTULTULTU

LTULTULTULTULTU

LTULTU

LUX

LUXLUXLUXLUXLUXLUXLUXLUXLUX

LUX

LUX

LUX

MKDMKDMKDMKD

MKD

MKD

MDG

MYS

MYSMYSMYSMYS

MLIMLI

MLTMLTMLTMLT

MLTMLT

MLTMLT

MLTMLT

MRTMRT

MEXMEX

MEXMEX

MEX

MEXMEXMEX

MDA

MDA

MDAMDAMDA

MNGMNG

MNGMNGMNG

MNEMNE

MNE

MNEMNEMNEMNEMNE

MOZ

MMRNPL

NLDNLDNLDNLDNLDNLDNLDNLD

NLDNLD

NLD

NLD

NICNICNIC

NERNER

NER

NGA

NOR NOR

NORNOR

PAKPAK

PAKPAK PAK

PAN

PAN

PAN

PANPANPANPAN

PAN

PANPANPAN

PRY

PRY

PRYPRYPRY

PRY

PRY

PRY

PRY

PRYPRYPRYPER

PER

PER

PERPER

PER

PER

PER

PER

PER

PERPERPERPER

PRTPRTPRTPRTPRT

PRTPRTPRTPRT

PRT

PRTPRTPRTROU

ROUROUROUROUROUROUROUROUROU

RUS

RUS

RUSRUS

RUSRUS

RWASEN

SLE

SVKSVKSVK

SVKSVKSVKSVK

SVKSVK

SVKSVK

SVN

SVN

SVN

SVN

ZAF

ESPESP

ESPESP

ESPESPESPESPESPESPESPESPESP

LKALKA LKALKASDN

SWESWESWESWESWESWESWESWESWESWE

SWE

SWE

SWE

CHECHECHECHECHE

CHE

CHE

CHE

TJKTJK

TJK

TZATZA

THATHATHATHA

THA

THA

THATHA

THA

THA

TGOTGOTGO

TUN TUR

TUR

TURTUR

TURTUR

TURTUR

TURTURTUR

TUR

TUR

TURTUR

UGA

UGAUGA

UGA

UKR

UKR

GBRGBRGBRGBRGBR

GBRGBRGBRGBR

GBR

GBR

GBRUSA

USAUSAUSAUSA

URY

URYURY

URY

URY

URYURYURYURY

UZB

VENVEN

VENVEN

VNM

VNMVNM

VNMVNM

VNMVNM

YEM

ZMBZMB

ZWE

24

68

10

25 35 45 55 65GINI Index (World Bank Estimate)

Happiness Score Fitted values


5

The murder rate is clearly positively correlated with income inequality (fig. 6), but the

suicide rate is correlated negatively, if at all (fig. 7).


Happiness), 0-to-10 scale; and murder rate per 100,000 inhabitants in 2000-18

Source: WHO; World Database on Happiness.

Figure 4: Happiness index in 2018 and suicide rates in 2016

AFG

AFG

AFGAFG

ALB

ALBALB

ALB

ALBALB

ALBALBALB

DZA

DZA

DZADZADZADZADZA

DZA

DZAAGO

AGO

ARGARG

ARM

ARMARM

ARMARMARMARMARMARMARMARM

AUSAUSAUSAUSAUSAUSAUSAUS

AUSAUSAUS

AUTAUTAUTAUTAUTAUTAUTAUTAUTAUTAUTAUTAUT

AUTAUTAUTAUT

AZEAZEAZE

AZE

AZEAZE

AZEAZE

BHRBHR

BHRBHRBHR

BHR

BGDBGDBGD

BGDBGDBGDBGDBGDBGDBGDBGD

BLR

BLRBLR

BELBELBELBEL

BELBELBELBELBELBELBELBELBELBELBELBEL

BLZ

BLZ

BLZBLZ

BEN

BTNBTNBTNBOL

BOL

BOL

BOL

BOL

BOL

BOL

BOLBOLBOL

BIHBIHBIHBIHBIHBIHBIH

BIHBIHBIH

BWA

BWA

BWA

BRA

BRABRA

BRABRABRABRABRABRABRABRABRA

BGRBGRBGRBGRBGR

BGRBGRBGRBGRBGRBGRBGR

BGR

BGR

BGRBGR

BFABFABFABFA

BFA

BFABFA

BFABFABDIBDIBDIBDI

BDIBDIBDI

KHM

KHMKHM

KHMKHMKHMCMR

CMRCMRCMRCMR

CAN

CANCANCAN

CAF

TCDTCD

CHLCHL

CHL

CHL

CHL

CHL

CHL

CHL

CHLCHLCHL

CHN

CHNCHN

CHN

CHNCHN

CHNCHN

CHNCHNCHNCHNCHN

COLCOL

COL

COLCOLCOLCOLCOLCOL

COL

COL

COLCOLCOL

COMCOM

COD

COG

CRICRICRICRICRI

CRICRI

CRI

CRICRI

CRI

CRICRICRICRI

CIV

HRVHRV

HRVHRVHRV

HRVHRVHRVHRVHRVCUB

CYPCYP

CYP

CYP

CYPCYPCYPCYPCYPCYPCYPCYPCYPCYP

CYPCYP

CZECZECZECZECZECZECZECZECZECZECZECZECZECZECZECZE

DNKDNKDNKDNKDNKDNKDNKDNKDNKDNKDNKDNKDNK

DNK

DNK

DNKDNK

DJI

DJI

DOM

DOM

DOM

DOM

DOM

DOM

DOM

DOMDOM

ECU

ECUECU

ECUECU

ECU

ECU

ECU

ECU

ECU

ECU

ECUECUECUECU

EGYEGY

EGY

EGY

EGY

EGYEGY

SLV

SLV

SLV SLV

SLV

SLV

SLV

SLV

SLV

SLV

SLV

SLVSLVSLV

ESTESTEST

ESTESTESTESTEST

ESTESTESTESTEST

EST

EST

ETHETH

FINFINFINFIN

FINFINFINFINFIN

FINFINFINFINFINFINFINFIN

FRAFRAFRAFRAFRAFRAFRAFRAFRA

FRA

FRAFRAFRAFRAFRAFRAFRA

GAB

GABGEOGEO

GEO

GEOGEO

GEOGEO

DEUDEUDEUDEU

DEUDEUDEUDEUDEUDEUDEUDEUDEUDEUDEUDEUDEU

GHAGHA

GHAGHA

GHA

GHA

GRCGRCGRCGRCGRCGRCGRCGRC

GRC

GRCGRC

GRC

GRC

GRC

GRCGRC

GTM

GTM

GTM

GTM

GTM

GTM

GTM

GTM

GTM

GTM

GTM

GTMGTM

GTM

GIN

GIN

GUY

GUY

GUY

HTIHTI

HNDHND

HND

HND

HND

HND

HND

HND

HND

HND

HND

HND

HNDHNDHND

HKGHKGHKGHKG

HKGHKGHUNHUNHUNHUNHUNHUNHUNHUNHUNHUNHUNHUNHUN

HUN

HUNHUN

ISLISLISL

IND

INDINDIND

IND

IND

INDINDINDINDINDIND

IDN

IDNIDNIDNIDNIDNIDNIDNIDN

IRN IRQIRQIRQ

IRQIRQ

IRLIRLIRLIRL

IRLIRLIRLIRLIRLIRLIRLIRLIRLIRLIRLIRLIRL

ISR

ISR

ISRISRISRISRISRISRISR

ITAITAITAITAITAITAITAITA

ITAITAITAITA

ITA

ITA

ITAITAITA

JAM JAMJAM

JAMJAM

JAMJPNJPNJPNJPNJPNJPNJPNJPNJPNJPNJPNJPNJPNJPNJPN

JORJOR

JOR

JOR

JOR

JORJORJORJORJORJORJOR

KAZ

KAZKAZKAZKAZKAZ

KAZKAZ

KEN

KENKENKENKENKENKEN

KEN

KENKEN

KOR

KORKORKORKORKOR

KOS

KOS

KOSKOS

KOS

KOSKOS

KOSKOS

KWTKWT

KWTKWT

KWTKWT

KGZKGZKGZ

KGZ KGZKGZKGZ

KGZKGZKGZKGZ

LAO

LVALVALVALVA

LVALVALVALVALVALVA

LVALVA

LVA

LVA

LVA

LBN

LBNLBNLBN

LBNLBNLBNLBNLBN

LSOLSO LSO

LBR

LBRLBRLBR

LBY

LBY

LTU

LTULTULTULTULTULTULTULTULTULTU

LTULTU

LTULTULTU

LUXLUXLUXLUX

LUXLUXLUXLUXLUXLUXLUXLUXLUXLUX

LUX

MKDMKDMKDMKDMKDMKD

MKD

MKD

MDG

MYSMYS

MYS

MYSMYSMYSMYSMYS

MLIMLI

MLT

MLTMLT

MLTMLT

MLTMLTMLTMLT

MLTMLT

MLTMLT

MLTMLT

MRTMRT

MUSMUSMUSMUSMUS

MEXMEXMEX

MEXMEXMEXMEX

MEXMEX

MEX

MEX

MEX

MDA

MDA

MDA

MNGMNGMNG

MNGMNGMNGMNGMNGMNGMNEMNE

MNE

MNEMNEMNEMNEMNEMNEMNE

MAR

MAR

MARMARMARMARMARMARMAR

MOZMOZMOZMOZ

MMRMMR

MMRMMRMMR

MMRMMR

NAMNPLNPLNPL

NPL

NPL

NPL

NPLNPL

NPLNPL

NLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLD

NLD

NLDNLD

NZL

NZL

NZL

NZL

NICNIC

NIC

NICNIC

NIC

NIC

NIC

NIC

NIC

NICNIC

NICNIC

NIC

NER

NER

NGA

NGA

NORNORNORNOR

NORNORNOR

OMNOMN

PAKPAK

PAK

PAK

PAK

PAK

PAKPAKPAKPAK

PAKPAK

PAN

PAN

PAN

PANPANPAN

PANPAN

PANPANPAN

PRY PRY

PRY

PRYPRYPRY

PRY

PRY

PRY

PRY

PRYPRYPERPERPERPER

PHLPHLPHLPHL

PHL

PHL

PHLPHLPHLPOL

POLPOLPOLPOLPOLPOLPOLPOLPOLPOLPOLPOLPRTPRTPRTPRTPRTPRTPRTPRT

PRTPRTPRTPRT

PRT

PRTPRTPRTPRT

PRI

QATQAT

QATQATQATQATQAT

ROU

ROU

ROU

ROU

ROUROUROUROUROUROUROUROUROUROUROUROU

ROURUS

RUS

RUSRUSRUS

RWARWA

RWA

RWARWARWA

SAUSAUSAU

SEN

SEN

SRB

SRB

SRBSRBSRB

SLESLE

SLE

SLE

SLESLE

SGPSGP

SGPSGPSGP

SVKSVKSVKSVKSVK

SVKSVKSVKSVKSVKSVKSVKSVKSVKSVKSVK

SVNSVNSVNSVNSVNSVNSVNSVNSVNSVNSVNSVN

SVN

SVN

SVNSVN

SOM

SOM ZAF

ZAF

ZAF

ZAF

ZAF

ZAFZAFZAF

ESPESPESPESPESPESPESPESPESPESPESPESPESPESPESPESPESP

LKALKALKA

LKALKALKALKALKALKALKA

SDN

SWESWESWESWESWESWESWESWESWESWESWESWESWESWE

SWE

SWESWE

CHECHECHECHECHECHECHECHECHECHE

CHE

CHE

CHECHE

SYRSYR

SYR

TWN

TWN

TWNTWNTWNTWN

TJKTJK

TJK

TJKTJKTJK

TZA

TZATZATZATZATZA

THATHATHATHA

THA

THA

THATHA

THA

THATHA

TGO

TTOTTO

TUNTUNTUN

TUN

TUR

TURTURTURTUR

TURTUR

TURTURTUR

UGA

UGAUGAUGA

UGA

UGA

UGAUGA

UGA

UKRUKR

AREAREARE

GBRGBRGBRGBRGBRGBRGBRGBRGBR

GBRGBRGBRGBR

GBR

GBR

GBRGBRUSAUSAUSA

USAUSA

USAUSAUSAUSAUSAUSAUSAUSAUSAUSAUSAURY

URYURY

URY

URYURY

URYURY

URY

URY

URYURYURYURY

UZB

UZBUZB

VEN VENVEN

VEN VENVENVEN

VENVEN

VEN

VEN

VNM

VNMVNMVNMVNMVNM

VNM

YEMYEM

YEM

YEM

YEMYEM

ZMB

ZMBZMBZMBZMB

ZMB

ZMB

ZWE

ZWEZWE

24

68

10

0 20 40 60 80 100Intentional homicide rate (Wikipedia)

Happiness Score Fitted values


6

Figure 5: Happiness index and suicide rates in 2000-18

Source: WHO; World Happiness Database.

Figure 6: Murder rate per 100,000 inhabitants and Gini coefficient of income

distribution in percentage terms in 2000-18

Afghanistan

Albania

Algeria

Angola

Argentina

Armenia

AustraliaAustria

Azerbaijan

Bahrain

Bangladesh

Belarus

Belgium

Belize

Benin

Bhutan

Bolivia

Bosnia and Herzegovina

Botswana

Brazil

Bulgaria

Burkina Faso

Burundi

Cambodia

Cameroon

Canada

Central African Republic

Chad

Chile

China

Colombia

Congo (Brazzaville)

Congo (Kinshasa)

Costa Rica

Croatia

Cyprus

Czech Republic

Denmark

Dominican Republic

Ecuador

Egypt

El Salvador

Estonia

Ethiopia

Finland

France

Gabon

Georgia

Germany

Ghana

Greece

Guatemala

Guinea

Haiti

HondurasHungary

Iceland

India

Indonesia

Iran

Iraq

IrelandIsrael

ItalyJamaica Japan

Jordan

Kazakhstan

Kenya

Kuwait

Kyrgyzstan

Laos

Latvia

Lebanon

Lesotho

Liberia

Libya

Lithuania

Luxembourg

Macedonia

MadagascarMalawi

Malaysia

Mali

Malta

Mauritania

Mauritius

Mexico

Moldova

MongoliaMontenegroMorocco

MozambiqueMyanmar

Namibia

Nepal

NetherlandsNew Zealand

Nicaragua

Niger

Nigeria

Norway

Pakistan

Panama

ParaguayPeruPhilippines

Poland

Portugal

Qatar

RomaniaRussia

Rwanda

Saudi Arabia

Senegal

Serbia

Sierra Leone

SingaporeSlovakia

Slovenia

Somalia

South Africa

South Korea

South Sudan

Spain

Sri Lanka

Sudan

SwedenSwitzerland

Syria

Tajikistan

Tanzania

Thailand

Togo

Trinidad & Tobago

Tunisia

TurkeyTurkmenistan

Uganda Ukraine

United Arab EmiratesUnited KingdomUnited States

Uruguay

Uzbekistan

Venezuela

Vietnam

Yemen

Zambia

Zimbabwe

34

56

78

0 10 20 30 40Suicide rate per 100,000 inhabitants in 2016

Happiness score in 2018 Fitted values

AFG

AFGAFG

ALB

ALBALB

DZADZA

DZA

AGOAGO

ARGARGARGARG

ARMARMARM

AUSAUS

AUTAUT

AUTAUT

AZE

AZEAZE

BHRBHRBHR

BGDBGDBGD

BLR

BLRBLR

BELBELBELBEL

BLZ

BLZ

BENBEN

BTNBTN

BOLBOL

BOLBOL

BIHBIHBIH

BWA

BWA

BWA

BRABRABRA

BGRBGRBGRBFA

BFABFA

BDIBDI

KHMKHMKHM

CMR

CMRCMR

CAN

CANCAN

CAFCAF

CAF

TCDTCDTCD

CHL

CHL

CHLCHL

CHN

CHNCHN

COL

COL

COLCOL

COMCOMCOM

CODCODCOGCOG

CRI

CRI

CRICRI

CIVCIV

HRVHRV

HRV

CYP

CYPCYP

CZE

CZECZE

DNKDNK

DNKDNK

DJI

DJI

DOM

DOMDOM

ECU

ECU

ECUECU

EGY

EGYEGY

SLVSLV

SLVSLVEST

ESTEST

ETHETHETH

FIN

FINFINFIN

FRAFRAFRAFRA

GAB

GABGABGEO

GEOGEO

DEUDEUDEU

DEU

GHAGHA

GHA

GRC

GRCGRCGRC

GTMGTM

GTMGTM

GIN

GINGIN

GUY

HTIHTI

HTI

HNDHND

HNDHND HUNHUN

HUN

ISLISL

IND

INDIND

IDNIDNIDN

IRNIRNIRQ

IRQIRQ

IRLIRLIRLIRL

ISRISRISR

ITAITAITAITA

JAM

JAMJAM

JPNJPNJPN

JOR

JORJORKAZ

KAZKAZ

KENKENKEN

KORKORKOR

KWT

KWTKWT

KGZKGZKGZLAOLAO

LVALVA

LVA

LBNLBNLBN

LSO

LSO

LBR

LBR

LBR

LBY

LBYLBY

LTU

LTULTU

LUXLUX

LUXLUX

MKDMKDMKD

MDGMDG

MWIMWI

MYSMYSMYS

MLIMLIMLI

MLT

MLT

MLT

MRT

MRTMRT

MUSMUS

MEX

MEX

MEX

MDA

MDAMDA

MNGMNGMNG

MNE

MNEMNE

MAR

MARMAR

MOZ

MMR

MMRMMR

NAM

NAMNPL

NPLNPL

NLDNLD

NLDNLDNZLNZL

NIC

NIC

NICNIC

NERNERNER

NGA

NGA

NGA

NORNOR

OMN

PAK

PAKPAK

PAN

PANPANPRY

PRY

PRY

PRYPER

PER

PERPER

PHLPHL

POLPOLPOLPRT

PRT

PRTPRT

QATQATQAT

ROU

ROU

ROU

ROURUS

RUS

RWARWA

SAUSAUSAU

SEN

SEN

SEN

SRBSRB

SLE

SLESLE

SGPSGP

SVKSVKSVK

SVN

SVNSVN

SOM

SOMSOM

ZAFZAF

ESPESPESPESP

LKALKALKASDN

SDN

SDN

SURSURSUR

SWESWE

SWESWE

CHECHE

CHECHE

SYR

SYRSYR

TJK

TJKTJK

TZA

TZATZA

THATHATHA

TGO

TGO

TTOTTO

TUN

TUNTUN

TUR

TURTURTKMTKM

UGAUGAUGA

UKR

UKR

ARE

ARE

GBRGBR

GBRGBR

USAUSAUSAURY

URY

URYURY

UZB

UZBUZB

VENVEN

VEN

VEN

VNMVNMVNM

YEM

YEM

YEM

ZMB

ZMB ZWE ZWE

ZWE

24

68

0 10 20 30 40 50Age-standardized suicide rates (WHO)

Fitted values Happiness Score


7

Figure 7: Suicide rate per 100,000 inhabitants and GINI coefficient of income

inequality in percentage terms in 2000-18

Source: WHO, WDI.

Wealth inequalities, especially inequalities at the very top of the wealth pyramid – the

share of wealth belonging to billionaires and millionaires – are positively correlated with

happiness indices (fig. 8-10).

ALBALBALBALB

DZA

ARGARGARMARMARM ARMARMARMARMARMARMARMARMARMARMARMARMARMAUSAUSAUSAUSAUTAUTAUTAUTAUTAUTAUTAUTAUTAUTAUTAUTAUT

AZEAZEAZEAZEAZE BGDBGDBGD

BLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBELBELBELBELBELBELBELBELBELBELBELBELBELBEN

BTNBTNBTN BOLBOL

BOLBOLBOLBOLBOL

BOLBOLBOLBOL

BIH BIHBIHBIHBIH

BWABWA

BRABRABRABRABRABRABRABRABRABRABRABRABRABRA

BGRBGRBGRBGRBGRBGRBGRBGRBGRBFABDI CMR

CANCANCANCANCANCPV

CHLCHLCHLCHLCHLCHLCHNCHN

COLCOLCOL

COL

COL

COL

COLCOLCOLCOLCOLCOLCOLCOLCOL

COGCRI CRICRICRICRICRICRICRI

CRICRICRICRICRICRICRICRICRICIV

HRVHRVHRVHRVHRVHRVHRVCYPCYPCYPCYPCYPCYPCYPCYPCYP CYPCYPCYPCZECZECZECZECZECZECZECZECZECZECZECZEDNKDNKDNKDNKDNKDNKDNKDNKDNKDNKDNKDNKDNK

DOMDOMDOM

DOMDOMDOMDOMDOM

DOMDOMDOMDOMDOM

DOMDOMDOM ECUECU

ECUECUECUECUECUECUECU

ECUECUECU

ECUECUECU

EGYEGYEGYEGY

SLVSLV

SLV

SLV

SLVSLVSLV

SLV

SLV

SLV

SLV

SLV

SLVSLV

SLV

SLV

SLV

ESTESTESTESTESTESTESTESTESTESTESTESTEST

ETH ETH

FJIFJIFINFINFINFINFINFINFINFINFINFINFINFINFIN FRAFRAFRAFRA FRAFRAFRAFRAFRAFRAFRAFRAFRA

GABGMBGMBGEOGEOGEOGEOGEO

GEOGEOGEOGEOGEOGEOGEOGEODEUDEUDEUDEUDEUDEUDEUDEUDEU GHAGRCGRCGRCGRCGRCGRCGRCGRCGRCGRCGRCGRCGRC

GTM

GTM

GTM

GNBHTI

HNDHNDHND

HND

HNDHND

HND

HND

HND

HND

HNDHND

HND

HND

HND

HUN HUNHUNHUNHUNHUNHUNHUNHUNHUNHUNHUNISLISLISLISLISL ISLISLISLISLISLISL ISL

INDIDN

IRNIRNIRN

IRQ

IRLIRLIRLIRLIRLIRLIRLIRLIRLIRLIRLIRLIRLISR ISRISRISRISRITAITAITAITAITAITAITAITAITAITAITAITAITA

JAM

JAM

JPNJORJORJOR

KAZKAZKAZKAZKAZKAZ

KAZKAZKAZKAZKAZKAZ KENKENKORKOSKOS

KOSKOSKOSKOSKOSKOS

KGZKGZKGZKGZ KGZKGZ KGZKGZKGZKGZ

KGZ

KGZKGZKGZKGZKGZKGZ

LVALVALVALVALVALVALVALVALVALVALVALVALBN

LSO

LBR

LTULTULTULTULTULTULTULTU LTULTULTULTU

LUXLUXLUXLUXLUXLUXLUXLUXLUXLUXLUXLUX MKDMKDMKDMKDMKDMKD

MDGMDG

MWI MWIMYSMYSMYSMYSMYSMDVMDVMLTMLTMLTMLTMLTMLTMLTMLTMLTMLT

MRT

MUSMUS

MEXMEXMEXMEXMEXMEX

MEXMEXMEXMEX

FSM

MDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDA MCOMCO

MNGMNGMNGMNG

MNGMNGMNEMNEMNEMNEMNEMNEMNEMNEMNEMNE MOZMOZ

MMR

NAM

NPLNPLNLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLD

NICNICNIC

NICNER

NOR NORNORNORNORNORNORNORNORNORNORNOR

PAKPAKPAKPAKPAKPAKPAKPAK

PANPANPANPANPANPANPANPAN

PANPANPANPANPANPAN

PANPANPAN

PRYPRYPRYPRY

PRYPRYPRYPRYPRYPRYPRYPRYPRY PRYPRY

PERPERPERPERPERPER

PRTPRTPRTPRTPRTPRTPRTPRTPRTPRTPRTPRTPRT ROUROUROUROUROUROUROUROUROUROU

RUSRUSRUSRUSRUS

RUSRUSRUSRUSRUSRUSRUSRUS

RWARWASTP

SEN

SYC

SLESVKSVKSVKSVKSVKSVKSVKSVKSVKSVKSVKSVKSVNSVNSVNSVNSVNSVNSVNSVNSVNSVNSVNSVNSLB

ZAF

ZAFZAFZAFZAF

ESPESPESPESPESPESPESPESPESPESPESPESPESP

LKALKA

LKALKASWESWESWESWESWESWESWESWESWESWESWESWESWE CHECHECHECHECHECHECHECHECHECHE SYRTJKTJKTJKTJK

TZA THATHATHATHATHATHATHATHATHATHATHATHATHATLS

TGOTONTONTUNTUN TURTURTURTURTURTURTURTURTURTUR

UGAUGAUGAUKRUKRUKRUKRUKRUKRUKRUKRUKRUKRUKR

GBRGBRGBRGBRGBRGBRGBRGBRGBRGBRGBRGBRUSAUSAUSAUSAUSAUSA URYURYURYURYURYURYURYURYURYURYURY

UZBUZBUZBVUT

VEN

VEN

VEN

VENVEN

VEN

VNMVNMVNMVNM VNMYEM ZMBZMB

02

04

06

08

01

00

20 30 40 50 60 70GINI Index (World Bank Estimate)

Intentional homicide rate 95% CI

Fitted values

AGO

ARGARGARG

ARMARMARM

AUSAUTAUT

BGDBGDBGD

BLR

BLR

BLRBLR

BELBEL BEN

BOL

BOLBOL

BIH BRABGR

CANCAN CHLCHL

COLCOLCOLCOLCRICRICRICRI

CIV

HRVHRV

CYPCYP

CZECZE

DNKDNKDOM

DOMDOMDOM

ECUECUECU

ECU

EGYEGY

SLV

SLV

SLVSLVESTESTETH ETH

FIN

FINFRA

FRA

GMBGMB

GEOGEOGEOGEO

DEUDEUDEU

GRCGRC GTM

GNB

HNDHND

HUN

HUNISL

IRLIRL

ISRITAITA

JOR

KAZ

KAZ

KEN

KOR

KGZ

KGZ

KGZKGZ

LVALVA

LSO

LTU

LTU

LUXLUX

MKD

MDGMWIMYSMLTMLT MRT

MEXMEXMEX

MDAMDA

MDAMDA

MNG

MNG

MNE MMR

NAMNPL

NLDNLDNORNOR

PAK PAK

PANPAN

PANPAN

PRYPRYPRY

PERPERPERPER

PRTPRT

ROUROU

RUS

RUSRUSRWA

RWA

STPSTP

SVK

SVNSVN ZAF ZAF

ESPESP

LKASWESWE

CHECHE

TJK

TZA

THA

THATHA

TGO

TUNTUN

TUR TURTUR

UGA

UKR

UKRUKR

GBRGBR

USAUSA

USAURY

URYURY

UZB

VUTVNMVNM

ZMBZMB

01

02

03

04

05

0


Age-standardized suicide rates (WHO) 95% CI

predicted SuicideRate


8

It is important to note that inequalities at the very top – billionaire and millionaire

‘intensity’, the ratio of billionaire/millionaire wealth to GDP – are not precisely correlated with

general measures of inequalities like Gini coefficients. Whereas Gini coefficients for wealth

and income distribution are positively correlated with one another (fig. 12), and billionaire

intensity is positively correlated with Gini coefficients for wealth distribution (fig.13), the

correlation of billionaire intensity with general income-distribution Gini coefficients is

negative, if present at all (fig. 14). That is to say, there are countries with quite an even

distribution of income, but levels of high billionaire intensity in GDP; e.g., Scandinavian

countries.

Figure 8: Happiness index (0-to-10 scale) and billionaire wealth from the Global

Wealth Report as a percentage of PPP GDP in 2018

Figure 9: Happiness index (0-to-10 scale) and billionaire wealth from the Forbes list

as a percentage of PPP GDP in 2018

Argentina

AustraliaAustria

Belgium

Brazil

Canada

Chile

China

Colombia

Cyprus

Czech Republic

Denmark

Egypt

Finland

France

Germany

Greece Hong Kong SAR, China

Iceland

India

Indonesia

IrelandIsrael

ItalyJapanKazakhstan

Kuwait

Lebanon

MalaysiaMexico

Morocco


Nigeria

Norway

PeruPhilippines

Poland

Portugal

Russia

Saudi Arabia Singapore

South Africa

South Korea

Spain

SwedenSwitzerland

Taiwan Province of China

Thailand

Turkey

Ukraine


34

56

78

0 10 20 30 40Billionaires wealth from GWR in 2018 to PPP GDP in 2016, %



9

Figure 10: Happiness index (on a scale of 0 to 10) and net worth of millionaires

according to the Global Wealth Report as a percentage of PPP GDP

Figure 11: Happiness index (on a scale of 0 to 10) and Gini coefficient of wealth

distribution in 2018, percentage terms

Argentina

AustraliaAustria

Belgium

Brazil

Canada

Chile

China

Colombia

Cyprus

Czech Republic

Denmark

Egypt

Finland

France

Germany


Iceland

India

Indonesia

IrelandIsrael


Kuwait

Lebanon

MalaysiaMexico

Morocco


Nigeria

Northern Cyprus

Norway

Palestinian Territories

PeruPhilippines

Poland

Portugal

Russia

Saudi ArabiaSingapore

Somalia

South Africa

South Korea

South Sudan

Spain

SwedenSwitzerland


Thailand

Turkey

Ukraine


34

56

78

0 20 40 60Net wealth of billionaires as a % of GDP in 2018


Argentina

AustraliaAustria

Belgium

Brazil

Canada

Chile

China

Colombia

Cyprus

Czech Republic

Denmark

Egypt

Finland

France

Germany

GreeceHong Kong SAR, China

Iceland

India

Indonesia

IrelandIsrael


Kuwait

Lebanon

MalaysiaMexico

Morocco


Nigeria

Norway

PeruPhilippines

Poland

Portugal

Russia

Saudi Arabia Singapore

South Africa

South Korea

Spain

SwedenSwitzerland


Thailand

Turkey

Ukraine

United Arab Emirates United KingdomUnited States

34

56

78

0 50 100 150 200 250Ratio of millionaires wealth to GDP, %



10

Source: GWR; World Happiness Database; WDI; Forbes billionaires’ list.

Figure 12: Income and wealth inequalities, Gini coefficients in 2000-18, percentage

terms

Figure 13: Wealth of billionaires as a percentage of GDP and Gini index of wealth

distribution, in percentage terms, in 2018

Argentina

AustraliaAustria

Belgium

Brazil

Canada

Chile

China

Colombia

Cyprus

Czech Republic

Denmark

Egypt

Finland

France

Germany


Iceland

India

Indonesia

IrelandIsrael


Kuwait

Lebanon

MalaysiaMexico

Morocco


Nigeria

Norway

PeruPhilippines

Poland

Portugal

Russia

Saudi ArabiaSingapore

South Africa

South Korea

Spain

SwedenSwitzerland

Syria


Thailand

Turkey

Ukraine

United Arab EmiratesUnited Kingdom United States

34

56

78

50 60 70 80 90 100Gini index of wealth distribution in 2018, %



11

Figure 14: Gini coefficient of income distribution and ratio of billionaire wealth to PPP

GDP, percentage terms

Source: WDI; GWR.

There are at least two different types of income inequality, presented in the schema

below: the same Gini coefficients of income distribution can result from the concentration of

inequalities at the high end or the low end of the income pyramid. The graphical

AFGAFGAFGAFGAFGAFGAFGAFGAFGAFGAFGAFGAFGAFGAFGAFGAFGALBALBALBALBALBALBALBALBALBALBALBALBALBALBALBALBALBALBALB DZADZADZADZADZADZADZADZADZADZADZADZADZADZADZADZADZADZADZAAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGO ATGATGATGATGATGATGATGATGATGATGATGATGATGATGATGATGATGATGATGARGARGARGARGARGARGARGARGARGARGARGARG

ARGARGARGARGARGARGARGARMARMARMARMARMARMARMARMARMARMARMARMARMARMARMARMARMARMARMAUSAUSAUSAUS

AUSAUSAUSAUSAUSAUS

AUSAUSAUS

AUSAUSAUS

AUS

AUSAUS

AUTAUT

AUT

AUT

AUT

AUTAUT

AUT

AUTAUTAUTAUT

AUTAUTAUT

AUT

AUT

AUTAUT

AZEAZEAZEAZEAZEAZEAZEAZEAZEAZEAZEAZEAZEAZEAZEAZEAZEAZEAZE BHSBHSBHSBHSBHSBHSBHSBHSBHSBHSBHSBHSBHSBHSBHSBHSBHSBHSBHSBHRBHRBHRBHRBHRBHRBHRBHRBHRBHRBHRBHRBHRBHRBHRBHRBHRBHRBHRBGDBGDBGDBGDBGDBGDBGDBGDBGDBGDBGDBGDBGDBGDBGDBGDBGDBGDBGD BRBBRBBRBBRBBRBBRBBRBBRBBRBBRBBRBBRBBRBBRBBRBBRBBRBBRBBRBBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBELBELBELBELBEL

BELBELBEL

BELBELBELBELBEL

BELBELBELBEL

BELBEL

BLZBLZBLZ BLZBLZBLZBLZBLZBLZBLZBLZBLZBLZBLZBLZBLZBLZBLZBLZBENBENBENBENBENBENBENBENBENBENBENBENBENBENBENBENBENBENBEN BTNBTNBTNBTNBTNBTNBTNBTNBTNBTNBTNBTNBTNBTNBTNBTNBTNBTNBTN BOLBOLBOLBOLBOLBOLBOLBOLBOLBOLBOLBOLBOLBOLBOLBOLBOLBOLBOLBIHBIHBIHBIHBIHBIHBIHBIHBIHBIHBIHBIHBIHBIHBIHBIHBIHBIHBIH BWABWABWABWABWABWABWABWABWABWABWABWABWABWABWABWABWABWABWABRABRABRABRABRABRABRA

BRABRABRABRABRA

BRA

BRA

BRA

BRA

BRABRABRA

BRNBRNBRNBRNBRNBRNBRNBRNBRNBRNBRNBRNBRNBRNBRNBRNBRNBRNBRNBGRBGRBGRBGRBGRBGRBGRBGRBGRBGRBGRBGRBGRBGRBGRBGRBGRBGRBGR BFABFABFABFABFABFABFABFABFABFABFABFABFABFABFABFABFABFABFABDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDI KHMKHMKHMKHMKHMKHMKHMKHMKHMKHMKHMKHMKHMKHMKHMKHMKHMKHMKHMCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMR

CANCANCANCANCANCAN

CANCAN

CANCANCANCAN

CANCANCANCANCANCANCAN

CPVCPVCPVCPVCPVCPVCPVCPVCPVCPV CPVCPVCPVCPVCPVCPVCPVCPVCPV CAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDCHLCHL

CHL

CHLCHL

CHLCHLCHL

CHL

CHLCHLCHL

CHL

CHL

CHL

CHL

CHL

CHL

CHL

CHNCHNCHNCHNCHNCHNCHNCHN

CHNCHN

CHNCHNCHNCHN

CHNCHNCHNCHNCHN

COLCOLCOL

COLCOLCOLCOL

COL

COL

COLCOLCOLCOLCOLCOL

COLCOL

COLCOL

COMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCODCODCODCODCODCODCODCODCODCODCODCODCODCODCODCODCODCODCODCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCRICRICRI CRICRICRI CRICRICRICRICRICRICRICRICRICRICRICRI CRICIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVHRVHRVHRVHRVHRVHRVHRVHRVHRVHRVHRVHRVHRVHRVHRVHRVHRVHRVHRV CYPCYPCYPCYPCYPCYP CYP

CYP

CYP

CYP

CYP

CYP

CYPCYP

CYP

CYP

CZECZECZECZECZE

CZE

CZE

CZE

CZE

CZECZECZECZECZECZECZE

CZE

CZECZE

DNK

DNKDNK

DNK

DNK

DNK

DNK

DNK

DNK

DNKDNK

DNKDNKDNK

DNK

DNKDNKDNK

DNK

DJIDJIDJIDJIDJIDJIDJIDJIDJIDJI DJIDJIDJIDJIDJIDJIDJIDJIDJI DMADMADMADMADMADMADMADMADMADMADMADMADMADMADMADMADMADMADMADOMDOMDOMDOMDOMDOMDOMDOMDOMDOMDOMDOMDOMDOMDOMDOMDOMDOMDOMECUECUECU ECUECUECUECUECUECUECUECUECUECUECUECUECUECUECUECUEGYEGYEGYEGYEGYEGYEGY

EGY

EGYEGYEGYEGY

EGYEGYEGY

EGY

EGY EGYEGYSLVSLVSLV SLVSLVSLVSLVSLVSLVSLVSLVSLVSLVSLVSLVSLVSLVSLV SLV GNQGNQGNQGNQGNQGNQGNQGNQGNQGNQGNQGNQGNQGNQGNQGNQGNQGNQGNQERIERIERIERIERIERIERIERIERIERIERIERIERIERIERIERIERIERIERI ESTESTESTESTESTESTESTESTESTESTESTESTESTESTESTESTESTESTESTETHETHETHETHETHETHETHETHETHETHETHETHETHETHETHETHETHETHETH FJIFJIFJIFJIFJIFJIFJIFJIFJIFJIFJIFJIFJIFJIFJIFJIFJIFJIFJIFINFINFINFINFINFINFINFINFIN

FIN FINFINFINFINFINFIN

FIN

FINFIN

FRAFRAFRAFRAFRA

FRA

FRA

FRAFRA

FRAFRAFRAFRA

FRAFRAFRAFRA

FRAFRA

GABGABGABGABGABGABGABGABGABGABGABGABGABGABGABGABGABGABGAB GMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGEOGEOGEOGEOGEOGEOGEOGEOGEOGEOGEOGEOGEOGEOGEOGEOGEOGEOGEO

DEUDEUDEUDEUDEUDEUDEUDEUDEU

DEUDEU

DEU

DEU

DEU

DEU

DEUDEU

DEU

DEU

GHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHA

GRCGRCGRC

GRC

GRC

GRC

GRC

GRCGRC

GRC

GRCGRCGRC

GRCGRCGRC

GRCGRC

GRC

GRDGRDGRDGRDGRDGRDGRDGRDGRDGRDGRDGRDGRDGRDGRDGRDGRDGRDGRDGTMGTMGTMGTMGTMGTMGTMGTMGTMGTMGTMGTMGTMGTMGTMGTMGTMGTMGTMGINGINGINGINGINGINGINGINGINGINGINGINGINGINGINGINGINGINGINGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNB GUYGUYGUY GUYGUYGUYGUYGUYGUYGUYGUYGUYGUYGUYGUYGUYGUYGUYGUY HTIHTIHTI HTIHTIHTIHTIHTIHTIHTIHTIHTIHTIHTIHTIHTIHTIHTIHTIHNDHNDHNDHNDHNDHNDHNDHNDHNDHNDHNDHNDHNDHNDHNDHNDHNDHNDHND

HKG

HKGHKG

HKGHKGHKGHKG

HKG

HKG

HKG

HUNHUNHUNHUNHUNHUNHUNHUNHUNHUNHUNHUNHUNHUNHUNHUNHUNHUNHUNISLISLISLISL

ISL

ISL

ISL

ISL

ISLISLISLISLISLISLISL

ISL

ISLISLISL

INDINDINDINDINDIND

INDIND

IND

INDINDINDINDIND

INDINDINDINDIND

IDNIRNIRNIRNIRNIRNIRNIRNIRNIRNIRNIRNIRNIRNIRNIRNIRNIRNIRNIRNIRQIRQIRQIRQIRQIRQIRQIRQIRQIRQIRQIRQIRQIRQIRQIRQIRQIRQIRQIRLIRLIRLIRL

IRLIRL

IRLIRLIRL

IRL

IRL

IRL

IRL

IRL

IRL

IRLIRL

IRLIRLISR

ISRISR

ISRISR

ISR

ISR

ISR

ISRISR

ISR

ISR

ISRISR

ISRISRISRISR

ISR

ITAITAITAITAITAITAITAITA ITA

ITAITAITA

ITA

ITAITAITAITAITA

ITA

JAMJAMJAM JAMJAMJAMJAMJAMJAMJAMJAMJAMJAMJAMJAMJAMJAMJAMJAM

JPN JPNJPNJPNJPNJPNJPNJPNJPNJPNJPNJPNJPNJPNJPNJPNJPNJPNJPN

JORJORJORJORJORJORJORJORJORJORJORJORJORJORJORJORJORJORJORKAZKAZKAZKAZKAZKAZ

KAZKAZ

KAZKAZKAZKAZ

KAZKAZKAZ

KAZKAZKAZKAZ

KENKENKENKENKENKENKENKENKENKENKENKENKENKENKENKENKENKENKENKIRKIRKIRKIRKIRKIRKIRKIRKIRKIRKIRKIRKIRKIRKIRKIRKIRKIRKIRKORKORKORKORKORKORKORKOR

KORKORKORKORKORKORKOR

KORKORKOR

KOR

KOSKOSKOSKOSKOSKOSKOSKOSKOSKOSKOSKOSKOSKOSKOSKOSKOSKOSKOS

KWTKWT

KWT

KWTKWT

KWTKWTKWTKWT

KWT

KWTKWTKWTKWTKWTKWTKWTKWTKWT

KGZKGZKGZKGZKGZKGZKGZKGZKGZKGZKGZKGZKGZKGZKGZKGZKGZKGZKGZ LAOLAOLAOLAOLAOLAOLAOLAOLAOLAOLAOLAOLAOLAOLAOLAOLAOLAOLAOLVALVALVALVALVALVALVALVALVALVALVALVALVALVALVALVALVALVALVA

LBNLBNLBN

LBNLBNLBNLBN

LBN

LBN

LBN

LBN

LBNLBN

LBNLBN

LBN

LBN

LBNLBN

LSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLBRLBRLBRLBRLBRLBRLBRLBRLBRLBRLBRLBRLBRLBRLBRLBRLBRLBRLBR LBYLBYLBYLBYLBYLBYLBYLBYLBYLBYLBYLBYLBYLBYLBYLBYLBYLBYLBYLTULTULTULTULTULTULTULTULTULTULTULTULTULTULTULTULTULTULTU LUXLUXLUXLUXLUXLUXLUXLUXLUXLUXLUXLUXLUXLUXLUXLUXLUXLUXLUX MACMACMACMACMACMACMACMACMACMACMACMACMACMACMACMACMACMACMKDMKDMKDMKDMKDMKDMKDMKDMKDMKDMKDMKDMKDMKDMKDMKDMKDMKDMKD MDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWI

MYSMYSMYSMYSMYSMYSMYSMYS

MYS

MYSMYSMYSMYS

MYSMYS

MYSMYSMYSMYS

MDVMDVMDVMDVMDVMDVMDVMDVMDVMDVMDVMDVMDVMDVMDVMDVMDVMDVMDVMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLTMLTMLTMLTMLTMLTMLTMLTMLTMLTMLTMLTMLTMLTMLTMLTMLTMLTMLT MHLMHLMHLMHLMHLMHLMHLMHLMHLMHLMHLMHLMHLMHLMHLMHLMHLMHLMHLMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMUSMUSMUSMUSMUSMUSMUSMUSMUSMUSMUSMUSMUSMUSMUSMUSMUSMUSMUS

MEXMEXMEX

MEXMEXMEXMEXMEX

MEXMEXMEXMEXMEXMEXMEXMEXMEXMEXMEX

FSMFSMFSMFSMFSMFSMFSMFSMFSMFSMFSMFSMFSMFSMFSMFSMFSMFSMFSMMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMNGMNGMNGMNGMNGMNGMNGMNGMNGMNGMNGMNGMNGMNGMNGMNGMNGMNGMNGMNEMNEMNEMNEMNEMNEMNEMNEMNEMNEMNEMNEMNEMNEMNEMNEMNEMNEMNE MARMARMARMARMARMARMARMARMARMARMAR

MARMARMARMARMARMARMARMAR

MOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMMRMMRMMRMMRMMRMMRMMRMMRMMRMMRMMRMMRMMRMMRMMRMMRMMRMMRMMR NAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNPLNPLNPLNPLNPLNPLNPLNPLNPLNPLNPLNPLNPLNPLNPLNPLNPLNPLNPL NLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLD

NZLNZL

NZLNZL

NZLNZLNZLNZL

NZL

NZLNZLNZLNZLNZLNZLNZLNZLNZLNZL

NICNICNIC NICNICNICNICNICNICNICNICNICNICNICNICNICNICNIC NICNERNERNERNERNERNERNERNERNERNERNERNERNERNERNERNERNERNERNER NGANGANGANGANGANGANGANGANGANGANGANGANGANGANGANGANGANGANGANOR

NORNOR

NORNORNORNOR

NOR

NOR

NORNORNOR

NOR

NOR

NOR

NORNORNORNOR

OMNOMNOMNOMNOMNOMNOMNOMNOMNOMNOMNOMNOMNOMNOMNOMNOMNOMNOMNPAKPAKPAKPAKPAKPAKPAKPAKPAKPAKPAKPAKPAKPAKPAKPAKPAKPAKPAK PLWPLWPLWPLWPLWPLWPLWPLWPLWPLWPLWPLWPLWPLWPLWPLWPLWPLWPLW PANPANPANPANPANPANPANPANPANPANPANPANPANPANPANPANPANPANPANPNGPNGPNGPNGPNGPNGPNGPNGPNGPNGPNGPNGPNGPNGPNGPNGPNGPNGPNG PRYPRYPRYPRYPRYPRYPRYPRYPRYPRYPRYPRYPRYPRYPRYPRYPRYPRYPRYPERPERPERPERPERPERPERPERPERPERPER

PER

PER

PER

PER

PERPER

PERPER

PHLPHL

PHLPHLPHLPHLPHLPHLPHLPHL

PHLPHL

PHL

PHL

PHL

PHLPHLPHLPHL

POLPOLPOLPOL POLPOLPOLPOL

POLPOL POLPOL POLPOLPOL POLPOL

POLPOLPRTPRTPRT

PRTPRTPRTPRT

PRT

PRTPRTPRTPRTPRTPRTPRTPRT

PRTPRTPRT

PRIPRIPRI PRIPRIPRI PRIPRIPRIPRIPRIPRIPRIPRIPRIPRIPRIPRI PRIQATQATQATQATQATQATQATQATQATQATQATQATQATQATQATQATQATQATQATROUROUROUROUROUROU ROUROUROUROUROUROUROUROUROUROUROUROUROU RUSRUS

RUS

RUSRUSRUS

RUS

RUS

RUS

RUS

RUS

RUSRUSRUS

RUSRUS

RUSRUS

RUS

RWARWARWARWARWARWARWARWARWARWARWARWARWARWARWARWARWARWARWAWSMWSMWSMWSMWSMWSMWSMWSMWSMWSMWSMWSMWSMWSMWSMWSMWSMWSMWSMSMRSMRSMRSMRSMRSMRSMRSMRSMRSMRSMRSMRSMRSMRSMRSMRSMRSMRSMRSTPSTPSTPSTPSTPSTPSTPSTPSTPSTPSTPSTPSTPSTPSTPSTPSTPSTPSTP

SAUSAUSAUSAU

SAU

SAU

SAU

SAU

SAUSAUSAUSAUSAUSAUSAU

SAUSAUSAUSAU

SENSENSENSENSENSENSENSENSENSENSENSENSENSENSENSENSENSENSENSRBSRBSRBSRBSRBSRBSRBSRBSRBSRBSRBSRBSRBSRBSRBSRBSRBSRBSRBSYCSYCSYCSYCSYCSYCSYCSYCSYCSYCSYCSYCSYCSYCSYCSYCSYCSYCSYCSLESLESLESLESLESLESLESLESLESLESLESLESLESLESLESLESLESLESLE

SGP

SGP

SGP

SGPSGPSGPSGP

SGP

SGP

SGP

SGPSGP

SGP

SGPSGPSGP

SGPSGP

SGP

SVKSVKSVKSVKSVKSVKSVKSVKSVKSVKSVKSVKSVKSVKSVKSVKSVKSVKSVK SVNSVNSVNSVNSVNSVNSVNSVNSVNSVNSVNSVNSVNSVN SVNSVNSVNSVNSVN SLBSLBSLBSLBSLBSLBSLBSLBSLBSLBSLBSLBSLBSLBSLBSLBSLBSLBSLBZAFZAFZAF

ZAFZAFZAFZAFZAF

ZAFZAF

ZAFZAF

ZAFZAFZAF

ZAFZAFZAF

ZAFESPESPESPESPESPESP

ESP

ESPESPESPESP

ESPESP

ESP ESPESPESPESP

LKALKALKALKALKALKALKALKALKALKALKALKALKALKALKALKALKALKALKA LCALCALCA LCALCALCALCALCALCALCALCALCALCALCALCALCALCALCA LCA VCTVCTVCTVCTVCTVCTVCTVCTVCTVCTVCTVCTVCTVCTVCTVCTVCTVCTVCTSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDN SURSURSURSURSURSURSURSURSURSURSURSURSURSURSURSURSURSURSURSWZSWZSWZSWZSWZSWZSWZSWZSWZSWZSWZSWZSWZSWZSWZSWZSWZSWZSWZ

SWE

SWESWE

SWESWE

SWE

SWE

SWE

SWE

SWESWESWE

SWE

SWE

SWE

SWE

SWE

SWE

SWECHE

CHE

CHE

CHE

CHECHE

CHE

CHE

CHE

CHE

CHE

CHECHE

CHE

CHE

CHECHE

CHECHE

SYRSYRSYRSYRSYRSYRSYRSYRSYRSYRSYR

TWNTWNTWN

TWNTWNTWNTWNTWN

TWN

TWN

TWN

TWNTWNTWNTWN

TWNTWNTWNTWN

TJKTJKTJKTJKTJKTJKTJKTJKTJKTJKTJKTJKTJKTJKTJKTJKTJKTJKTJKTZATZATZATZATZATZATZATZATZATZATZATZATZATZATZATZATZATZATZA THATHATHATHATHATHA

THATHATHATHA

THATHATHA THA

THA

THATHA

THATHA

TLS TLSTLSTLSTLSTLSTLSTLSTLSTLSTLSTLSTLSTLSTLSTLSTLSTLSTLS TGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTONTONTONTONTONTONTONTONTONTONTONTONTONTONTONTONTONTONTON TTOTTOTTO TTOTTOTTOTTOTTOTTOTTOTTOTTOTTOTTOTTOTTOTTOTTO TTOTUNTUNTUNTUNTUNTUNTUNTUNTUNTUNTUNTUNTUNTUNTUNTUNTUNTUNTUNTURTURTURTURTUR

TURTUR

TUR

TUR

TUR

TUR

TUR

TUR

TURTUR

TURTURTURTUR

TKMTKMTKMTKMTKMTKMTKMTKMTKMTKMTKMTKMTKMTKMTKMTKMTKMTKMTKM UGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUKRUKRUKRUKRUKRUKR

UKR

UKR

UKRUKRUKRUKRUKRUKR

UKR UKRUKRUKRUKR

AREAREAREAREAREARE

AREAREAREAREAREAREARE

AREAREAREAREAREARE

GBRGBRGBRGBRGBRGBRGBRGBR

GBRGBRGBRGBRGBRGBRGBRGBRGBRGBRGBR

USA

USAUSA

USAUSAUSAUSAUSA

USAUSAUSAUSAUSAUSAUSAUSAUSAUSAUSA

URYURYURY URYURYURYURYURYURYURYURYURYURYURYURYURYURYURYURYUZBUZBUZBUZBUZBUZBUZBUZBUZBUZBUZBUZBUZBUZBUZBUZBUZBUZBUZB VUTVUTVUTVUTVUTVUTVUTVUTVUTVUTVUTVUTVUTVUTVUTVUTVUTVUTVUT VENVENVEN VENVENVENVENVENVENVENVENVENVENVENVENVENVENVENVENVNMVNMVNMVNMVNMVNMVNMVNMVNMVNMVNMVNMVNMVNMVNMVNMVNMVNMVNM YEMYEMYEMYEMYEMYEMYEMYEMYEMYEMYEMYEMYEMYEMYEMYEMYEMYEMYEMZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWE05

10

15

20

40 60 80 100GINI Index (wealth distribution)

Wealth of billionares as a % of PPP GDP in 2018 Fitted values

ALBALBALBALBDZA AGOAGOARGARGARGARGARGARGARGARGARGARGARGARG

ARGARGARGARGARMARMARM ARMARMARMARMARMARMARMARMARMARMARMARMARMAUSAUSAUS

AUSAUT

AUT

AUTAUT

AUT

AUTAUTAUT

AUT

AUTAUTAUT

AUT

AZEAZEAZEAZEAZE BGDBGDBGDBGDBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBLRBELBEL

BELBELBEL

BELBELBELBELBEL

BELBELBEL

BEN BEN BENBTNBTNBTNBTN BOLBOLBOLBOL BOLBOLBOLBOLBOLBOLBOLBOLBOLBOLBOLBIH BIHBIHBIHBIH BWABWABRABRABRABRABRABRA

BRABRABRA

BRABRA

BRA

BRA

BRA

BGRBGRBGRBGRBGRBGRBGRBGRBGRBFABDI BDI CMRCMR CMR

CANCAN

CAN

CANCAN

CPVCPVCAF CAFTCD TCDCHL

CHL

CHLCHLCHL

CHL

CHL

CHN

CHN

COLCOLCOL

COLCOL

COL

COL

COLCOLCOLCOLCOLCOL

COLCOL

COMCOMCODCOD COGCOGCRI CRICRICRICRICRICRICRICRICRICRICRICRICRICRICRICRICIVCIVCIVHRVHRVHRVHRVHRVHRVHRVCYPCYPCYP

CYP

CYP

CYP

CYP

CYP

CYP CYP

CYP

CYP

CZE

CZE

CZE

CZE

CZE

CZECZECZECZECZE

CZECZE

DNK

DNK

DNK

DNK

DNK

DNK

DNKDNKDNKDNK

DNK

DNK

DNK

DJI DJIDJI DOMDOMDOMDOMDOMDOMDOMDOMDOMDOMDOMDOMDOMDOMDOMDOMDOM ECUECUECUECUECUECUECUECUECUECUECUECUECUECUECU

EGYEGYEGYEGYEGY

SLVSLVSLVSLVSLVSLVSLVSLVSLVSLVSLVSLVSLVSLVSLVSLVSLVESTESTESTESTESTESTESTESTESTESTESTESTESTETH ETH ETHFJI FJIFJIFINFINFINFINFINFINFINFINFIN

FINFINFINFIN

FRAFRA

FRA

FRA

FRAFRAFRAFRAFRAFRA

FRAFRAFRA

GABGAB GMBGMBGMB GEOGEOGEOGEOGEOGEOGEOGEOGEOGEOGEOGEOGEOGEOGEOGEOGEO

DEUDEU

DEUDEUDEUDEU

DEUDEU

DEU

GHAGHA

GRC

GRC

GRC

GRC

GRCGRC

GRC

GRCGRC

GRC

GRCGRCGRC

GTMGTMGTMGINGINGINGNB GNBHTI HNDHNDHNDHNDHNDHNDHNDHNDHND HNDHNDHNDHNDHNDHNDHUN HUNHUNHUNHUNHUNHUNHUNHUNHUNHUNHUNISL

ISL

ISL

ISL

ISL

ISLISLISLISLISLISL ISL

IND

IRNIRNIRNIRNIRNIRQIRQ

IRL

IRLIRL

IRLIRL

IRL

IRL

IRL

IRL

IRL

IRL

IRL

IRL

ISR

ISR

ISR

ISRISR

ITAITAITAITAITAITAITA

ITAITA

ITAITAITAITA

JAMJAMJPN JORJORJORJORKAZKAZKAZKAZ KAZ

KAZKAZ

KAZKAZKAZKAZKAZKAZ

KAZKAZ

KENKENKIR

KORKOR

KORKOR

KOSKOSKOSKOSKOSKOSKOSKOSKOSKOSKOS KGZKGZKGZKGZ KGZKGZ KGZKGZKGZKGZKGZKGZKGZKGZKGZKGZKGZ LAO LAOLAOLVA LVALVALVALVALVALVALVALVALVALVALVA

LBN

LSOLSOLBRLBR LTULTULTULTULTULTULTULTU LTULTULTULTULUXLUXLUXLUXLUXLUXLUXLUXLUXLUXLUXLUX LUX MKDMKDMKDMKDMKDMKD MDGMDGMDGMDGMWI MWI

MYSMYS

MYS

MYSMYS

MYS

MDVMDVMLIMLIMLIMLTMLTMLTMLTMLTMLTMLTMLTMLTMLT MRTMRTMRTMRT MUSMUS

MEXMEXMEXMEXMEX

MEXMEXMEXMEXMEXMEX

FSMFSMMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDAMDA MNGMNGMNGMNGMNGMNGMNGMNEMNEMNEMNEMNEMNEMNEMNEMNEMNE MOZMOZ MOZMMR NAMNAMNAMNPLNPLNLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLDNLD

NICNICNICNICNERNERNER NER NGA NGA

NOR NORNORNOR

NOR

NORNORNOR

NOR

NOR

NOR

NOR

PAKPAKPAKPAKPAKPAKPAK PAK PANPANPANPANPANPANPANPANPANPANPANPANPANPANPANPANPANPNG PRYPRYPRYPRYPRYPRYPRYPRYPRYPRYPRYPRYPRY PRYPRYPRYPERPERPERPERPERPERPERPERPERPERPER

PER

PER

PER

PER

PERPER

PRTPRT

PRTPRT

PRT

PRTPRTPRTPRTPRTPRTPRTPRT

ROUROUROUROUROUROUROUROUROUROU

RUSRUS

RUSRUSRUS

RUS

RUS

RUS

RUS

RUS

RUSRUSRUS

RUSRUS

RUS

RWA RWARWARWAWSMSTPSTP SENSENSEN SYCSLESLESVKSVKSVKSVKSVKSVKSVKSVKSVKSVKSVKSVKSVNSVNSVNSVNSVNSVNSVNSVNSVNSVNSVNSVN SLBSLBZAF

ZAFZAF

ZAFZAFESPESPESP

ESP

ESPESP

ESPESPESPESP

ESPESPESP

LKALKALKA LKALKASDN

SWESWE

SWE

SWE

SWE

SWE

SWESWESWE

SWE

SWE

SWE

SWE

CHE

CHE

CHE

CHE

CHE

CHECHE

CHE

CHE

CHE

SYRTJKTJKTJKTJK TJK TZA TZATZA THATHATHATHATHATHATHA

THATHATHATHA

THA

THA

TLSTLSTLS TGO TGOTGOTONTON TUNTUNTUN TURTUR

TUR

TURTUR

TUR

TUR

TUR

TUR

TUR

TUR

TURTUR

TURTUR

UGAUGAUGAUGAUGAUKRUKRUKRUKR

UKR

UKR

UKRUKRUKRUKRUKRUKR

UKRUKRUKR GBRGBRGBRGBRGBRGBRGBRGBRGBR

GBRGBRGBR

USAUSAUSA

USAUSAUSA

URYURYURYURYURYURYURYURYURYURYURYUZBUZBUZBVUT VENVENVENVENVENVENVNMVNMVNMVNM VNMVNMVNMVNMYEMYEM ZMB ZMBZMBZMBZMBZWE05

10

15

20


Wealth of billionares as a % of PPP GDP in 2018 Fitted values


12

interpretation of the Gini coefficient is the ratio of the area between the line of complete

equality and the standard Lorenz curve to the area of the shaded triangle.

In the first chart, the ABC triangle has about the same area as the ‘Lorenz curve–

complete equality’ area – and the same Gini coefficient – but inequality in concentrated at

the top end. So the 90% of the population at the poorer end of the income distribution are

totally equal in their income among themselves, but only account for 50% of total income,

whereas the richest 10% have the other 50% of total income; the per-capita income of the

rich is exactly nine times higher than the per capita income of the poor.4

In the second chart, the ABD triangle also has about the same area as the ‘Lorenz

curve–complete equality’ area – and the same Gini coefficient – but inequality exists

because the lower half of society has no income at all, and the other half has all the income.

The data seem to suggest that the first type of income distribution – ‘90% poor and

equal, 10% rich’ – is better for happiness than the second type.

Schema: Two types of inequality with the same Gini coefficient

0

10

20

30

40

50

60

70

80

90

100

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

Shar

e in

inco

me

, %

Share of population, %

Complete equality among the poor, but high income of the few rich

Share in income - standard LorenzcurveComplete equality

Share in income: 90% of poor havethe same income

C

B


13

An explanation of the ‘inequality–happiness’ relationship in terms of statics (space –

geography) and dynamics (time – history) could be the ‘big fish in a small pond’ effect. This

is a model developed by Marsh and Parker (1984) to explain why good students prefer to

stay in a class in which they are above the average level, rather than be in a more

challenging learning environment where they are below the average level. This effect can

be used to explain one of the paradoxes of happiness: Strong growth is usually

accompanied by growing income inequalities (Popov, 2018a, fig. 10), so rapid growth is

often associated with low happiness scores. A paper by Brockmann, Delhey, Welzel, and

Hao (2008) refers to the concept of “frustrated achievers” and explains the decline in

happiness scores in China through the deterioration of relative incomes for the majority of

the population due to rises in income inequality.

But there is a different relationship with regards to levels and change, stocks and

flows, and space and time dimensions: with low levels of inequality people feel unhappy –

the dream of the ‘big fish in a small pond’ is out of reach – but the transition to higher levels

of inequality, when the relative position of the majority deteriorates in relation to the average,

makes people even more unhappy temporarily, during the transition. When transition to a

higher level of inequality is over, people – maybe new generations – start to feel happier.


14

The other explanation could be a different relationship between happiness and inequality in

rich and poor countries. The dependence of happiness on income is characterised by a

rising but concave curve (fig. 14) that reaches its maximum at a level of income of about

$75,000 (the level of very rich Norway and Kuwait) and which increases only marginally after

the income level of about $30,000 (the level of the poorest OECD members – Greece, Chile,

and Estonia). Whereas happiness index scores rise from 3 to 6 with a rise of PPP GDP per

capita from less than $1,000 to $30,000, they only increase from 6 to 7 when per-capita

income rises from $30,000 to $75,000 (fig. 15).


Happiness), 0-to-10 scale, and PPP GDP per capita in 2018, Dollars

Source: World Database on Happiness, 2019; WDI.

It may well be that in rich countries, the ‘money can’t buy happiness’ story is more

true than in poor countries; i.e., a marginal increase in happiness due to a unit increase in

income in rich countries is lower than in poor countries.

Afghanistan

Albania

Algeria

Angola

Argentina

Armenia

AustraliaAustria

Azerbaijan

Bahrain

Bangladesh

Belarus

Belgium

Belize

Benin

Bhutan

Bolivia

Bosnia and Herzegovina

Botswana

Brazil

Bulgaria

Burkina Faso

Burundi

Cambodia

Cameroon

Canada

Central African Republic

Chad

Chile

China

Colombia

Congo (Brazzaville)

Congo (Kinshasa)

Costa Rica

Croatia

Cyprus

Czech Republic

Denmark

Dominican Republic

Ecuador

Egypt

El Salvador

Estonia

Ethiopia

Finland

France

Gabon

Georgia

Germany

Ghana

Greece

Guatemala

Guinea

Haiti

Honduras Hong Kong SAR, ChinaHungary

Iceland

India

Indonesia

Iran

Iraq

IrelandIsrael

Italy

Ivory Coast

Jamaica Japan

Jordan

Kazakhstan

Kenya

Kosovo

Kuwait

Kyrgyzstan

Laos

Latvia

Lebanon

Lesotho

Liberia

Libya

Lithuania

Luxembourg

Macedonia

MadagascarMalawi

Malaysia

Mali

Malta

Mauritania

Mauritius

Mexico

Moldova

MongoliaMontenegroMorocco

MozambiqueMyanmar

Namibia

Nepal


Nicaragua

Niger

Nigeria

Norway

Pakistan

Panama

ParaguayPeruPhilippines

Poland

Portugal

Qatar

RomaniaRussia

Rwanda

Saudi Arabia

Senegal

Serbia

Sierra Leone

SingaporeSlovakiaSlovenia

South Africa

South Korea

Spain

Sri Lanka

Sudan

SwedenSwitzerland


Tajikistan

Tanzania

Thailand

Togo

Trinidad & Tobago

Tunisia

TurkeyTurkmenistan

UgandaUkraine


Uruguay

Uzbekistan

Venezuela

Vietnam

Yemen

Zambia

Zimbabwe

34

56

78

0 50000 100000 150000PPP GDP per capita in 2018, dollars

Happiness score in 2018 95% CI

predicted happinessscore


15

First, for rich people, non-income determinants of happiness probably play a larger role, so

the negative effects of inequality are not counterweighted by higher incomes. Second, in

rich countries, people derive pleasure from being better off than most of the world population

– i.e., their ‘American dream’ has been achieved already – so inequality in their own country

is less important and has only negative consequences.

This paper uses the databases from the Forbes billionaires list and the Global Wealth

Report (GWR), as well as from the World Happiness Report and the World Database on

Happiness. I use the data to examine the relationship between income inequality, and

happiness for nearly 20 years (2000-18) for over 200 countries.

It turns out that inequality grows together with happiness in relatively poor countries,

but harms happiness in rich countries. On the other hand, inequality of wealth distribution –

wealth is a stock variable whereas income is a flow – is positively linked to happiness in all

countries. Wealth distribution at the very top of the pyramid – billionaire and millionaire

wealth as a proportion of GDP – is one of the most important determinants of happiness

index scores in all countries.

The stylised fact is that there are two ‘typical’ statistical portraits of a happy country.

One is of a country with relatively high income compared to the world average, a low level

of income inequality, but high wealth inequality and especially high wealth inequality at the

very top – millionaire and billionaire intensity. This is very much the image of a Scandinavian

country – high income compared to the rest of the world, low income inequality within the

country, but pretty high wealth inequality and billionaire intensity within the country. The

other type of happy country has lower levels of income, but high income and wealth

inequalities, especially at the top of the wealth pyramid. This is the Latin American and

African model; e.g., Honduras, Bolivia, Ecuador, Costa Rica, South Africa, and Zimbabwe.

Data

Income. Data on income are from the World Development Indicators database – purchasing

power parity (PPP) GDP per capita.

Happiness. Data on happiness come from the World Happiness Report, as well as

from the World Database on Happiness. This represents individuals’ self-perception of how

happy they are. The scale is from 0 to 10, and the estimates are derived from the Gallup

World Poll, the World Values Survey, and other sources.


16

Income and wealth inequalities. Income inequalities data are from the World

Development Indicators database (WDI) and derived from national household surveys of

income and consumption in various countries. Wealth inequalities are computed through

extrapolation: first, regressions between the components of personal financial and non-

financial wealth and its determinants (real consumption; population density; market

capitalisation rate; public pensions as a percentage of GDP; domestic credits available to

private sector; and Gini coefficient of income distribution) are computed for about 40

countries for which these data are available, then an extrapolation is made for countries that

do not have estimates of these components of personal wealth (Davies, Sandström,

Shorrocks, and Wolff, 2007).

Wealth of high net worth individuals (HNWI) – billionaires and millionaires.

Sample surveys tend to omit HNWIs, so income and wealth distribution at the very top of

the pyramid is very much underestimated. That is why this paper uses the Forbes list of

billionaires, which reports the wealth of all billionaires in the world since 1996 – see Popov

(2018a) for details – and the Credit Suisse Global Wealth Report, which makes a number of

adjustments to the Forbes data on billionaires (GWD, 2018, pp. 110-113) and estimates the

number and wealth of multi-millionaires, millionaires, and other HNWIs.5

Forbes data show a higher ratio of billionaire wealth to GDP than the GWR data. For

instance, for Hong Kong, the comparison is 58% and 30% respectively. But overall, these

two estimates are strongly correlated (fig. 16).

Figure 16: Billionaire intensity in percentage terms of PPP GDP according to the

Forbes list and according to the Global Wealth Report


17

Source: GWR; Forbes.

As figure 17 suggests, the ratio of millionaire wealth to GDP is correlated with the ratio of

billionaire wealth to GDP. However, data on the wealth of millionaires have the advantage

of including more countries. In 2018, there were only 24 countries, out of over 150, for which

data were available that did not have a single millionaire; in contrast the number of countries

without billionaires was nearly 100 out of over 150.

AfghanistanAlbaniaAlgeriaAngolaArgentinaArmenia

AustraliaAustria

AzerbaijanBahrainBangladeshBelarusBelgium

BelizeBeninBhutanBoliviaBosnia and HerzegovinaBotswana

Brazil

BulgariaBurkina FasoBurundiCambodiaCameroon

Canada

Central African RepublicChad

Chile

ChinaColombiaCongo (Brazzaville)Congo (Kinshasa)Costa RicaCroatia

Cyprus

Czech Republic

Denmark

Dominican RepublicEcuadorEgypt

El SalvadorEstoniaEthiopia

Finland

France

Gabon

GeorgiaGermany

GhanaGreece

GuatemalaGuineaHaitiHonduras

Hong Kong SAR, China

Hungary

Iceland

IndiaIndonesia

IranIraq

Ireland

Israel

Italy

Ivory CoastJamaicaJapan

JordanKazakhstan

KenyaKosovoKuwaitKyrgyzstanLaosLatvia

Lebanon

LesothoLiberiaLibyaLithuaniaLuxembourgMacedoniaMadagascarMalawi

Malaysia

MaliMaltaMauritaniaMauritius

Mexico

MoldovaMongoliaMontenegroMorocco

MozambiqueMyanmarNamibiaNepal


NicaraguaNiger

Nigeria

Norway

PakistanPanamaParaguayPeru

Philippines

PolandPortugal

QatarRomania

Russia

RwandaSaudi Arabia

SenegalSerbiaSierra Leone

Singapore

SlovakiaSlovenia

South AfricaSouth Korea

Spain

Sri LankaSudan

Sweden

Switzerland


TajikistanTanzania

Thailand

TogoTrinidad & TobagoTunisiaTurkey

TurkmenistanUgandaUkraineUnited Arab Emirates

United Kingdom

United States

UruguayUzbekistanVenezuelaVietnamYemenZambiaZimbabwe0

20

40

60

Net w

ea

lth o

f b

illio

nair

es a

s a

% o

f G

DP

in

201

8

0 10 20 30 40Billionaires wealth from GWR in 2018 to PPP GDP in 2016, %


18

Suicide and murder rates. These data come from World Health Organization

statistics on causes of death.6 Murder rates statistics is also available from the UNODC

(United Nations Office on Drugs and Crime), which collects statistics mostly from WHO but

from other sources as well.

Figure 17: Millionaire and billionaire intensity of PPP GDP according to the Global

Wealth Report, percentage terms

AfghanistanAlbaniaAlgeriaAngolaArgentinaArmenia

AustraliaAustria

AzerbaijanBahrainBangladeshBelarusBelgium

BelizeBeninBhutanBoliviaBosnia and HerzegovinaBotswanaBrazil

BulgariaBurkina FasoBurundiCambodiaCameroon

Canada

Central African RepublicChadChileChina

ColombiaCongo (Brazzaville)Congo (Kinshasa)Costa RicaCroatia

Cyprus

Czech Republic Denmark

Dominican RepublicEcuadorEgypt

El SalvadorEstoniaEthiopia

Finland

FranceGabonGeorgia

Germany

GhanaGreece

GuatemalaGuineaHaitiHonduras

Hong Kong SAR, China

Hungary

Iceland

IndiaIndonesia

IranIraq

Ireland

Israel

Italy

Ivory CoastJamaica JapanJordan

Kazakhstan

KenyaKosovo

Kuwait

KyrgyzstanLaosLatvia

Lebanon

LesothoLiberia LibyaLithuania LuxembourgMacedoniaMadagascarMalawi

Malaysia

Mali MaltaMauritaniaMauritiusMexico

MoldovaMongoliaMontenegroMoroccoMozambiqueMyanmarNamibiaNepalNetherlands

New Zealand

NicaraguaNigerNigeria

Norway

PakistanPanamaParaguay

PeruPhilippinesPolandPortugal

QatarRomania

Russia

RwandaSaudi Arabia

SenegalSerbiaSierra Leone

Singapore

SlovakiaSloveniaSouth Africa

South KoreaSpain

Sri LankaSudan

Sweden

Switzerland


TajikistanTanzania

Thailand

TogoTrinidad & TobagoTunisia

Turkey

TurkmenistanUgandaUkraine

United Arab Emirates United KingdomUnited States

UruguayUzbekistanVenezuelaVietnamYemenZambiaZimbabwe01

02

03

04

0

0 50 100 150 200 250Ratio of millionaires wealth to GDP, %

Billionaires wealth from GWR in 2018 to PPP GDP in 2016, % Fitted values


19

Source: Global Wealth Report.

Results: More billionaires – less inequality, fewer murders, and more happiness

To begin with, it is important to remember that billionaire intensity is negatively related to

income inequalities – the lower the level of inequality, the more billionaires per unit of GDP

the country has, even controlling for the level of income and for random effects.7 This

relationship is counterintuitive and suggests that inequalities at the very top are more

pronounced than inequalities among other income groups.

Table 1 reports the results of regressions of happiness indices on various

determinants without controls for fixed and random effects.

Models 1.1-1.6 link happiness indices to income and wealth inequalities and to

billionaire and millionaire intensity – with and without controls for per capita income – with

similar results: wealth inequality and billionaire and millionaire intensity have positive effect

on happiness, whereas income inequality affects happiness negatively. Models 1.1-1.2 are

based on cross-section data from about 150 countries; models 1.3-1.6 are based on panel

data. In models 1.3-1.6 (panel data) coefficients become insignificant, if controls for fixed

effects or random effects are introduced, which probably suggests non-linearity, so an

interaction term – per capita income multiplied by Gini coefficient of income distribution – is

introduced in the next model (1.7).


20

Table 1: Regression results of happiness indices on various determinants. Cross

country and panel data without control for random and fixed effects (T-statistics add

z- statistics in brackets)

Dependent variable – Happiness

Model,

parameters, N /

Variables

1.1, Cross-country,

N=139

1.2, Cross-

country,

N=151

1.3, Panel,

N=1967

1.4, Panel,

N=1967

1.5, Panel,

N=1967

1.6, Panel,

N=760

1.7, Panel,

N=760

PPP GDP per

capita, $, Ycap

.00002*** (5.2) 3.8e-06

(1.22)

3.6e-06

(1.3)

.00005***

(14.8)

.0001***

(10.2)

Gini coefficient of

income

distribution, %,

GINIincome

-.02**

(1.9)

.04***

(9.0)

.08***

(10.6)

Gini coefficient of

wealth

distribution, %,

GINIwealth

.03**

(2.4)

.02**

(2.0)

0.02***

(4.3)

.01***

(3.6)

Billionaires’

wealth to PPP

GDP ratio, %

Billionaires’Inte

nsity

0.13***

(5.7)

0.12***

(5.2)

0.03***

(9.6)

0.03**

(2.1)

Millionaires’

wealth to PPP

GDP ratio, %

Millionaires’Inte

nsity

.01***

(8.56)

.008***

(4.8)

.06***

(3.9)

Interaction term

(Ycap *

GINIincome)

-2.6e-06 ***

(-6.6)

Constant 3.9***

(5.6)

3.6***

(6.8)

5.4***

(162.9)

5.4***

(91.9)

3.2***

(13.6)

2.2***

(7.1)

1.3***

(4.0)

R2 , % 44 58 9 12 8 45 48

*, **, *** - denotes significance at 10, 5 and 1% respectively.

Model 1.7 is a specification with the threshold of per-capita income. The resulting

equation has the following form:

Happiness = 1.3*** + 0.0001Ycap*** + 0.01GINIwealth*** + 0.06Millionaires’Intensity***

+

+ 1.7e-06GINIincome*** (30700– Ycap***)


21

This implies that wealth inequality, both in general and at the top of the wealth pyramid

(‘millionaire intensity’), has a positive impact on happiness, whereas income inequality has

a positive impact only in countries with per capita income above $30,700.

Table 2 reports the results of regressions using panel data and controlling for random

and fixed effects. Better results are obtained by controlling for random effects; variation

within the groups – 19 years for the same country – is not enough in most cases to receive

meaningful results. But regressions with fixed-effects controls are reported as well, for

consistency.

Model 2.1 shows a significant and positive dependence of happiness indices on per-

capita income and income inequalities, but once wealth inequalities and billionaire intensity

variables are introduced into the right-hand side, coefficients become insignificant.

Model 2.2 suggests a non-linear relationship: Happiness = 3.7*** + 0.00009Ycap***

+1.6e-06GINIincome*** (25000 – Ycap***)

In countries with per-capita income (PPP GDP per capita) higher than roughly US$

25,000 for a particular period, income inequalities have a negative impact on happiness, but

in poorer countries the relationship is positive.

Model 2.3 shows the same non-linear relationship with a similar threshold of per-

capita income ($22,100 + 388*Billionaires’ intensity). The interesting twist here is that the

higher the level of billionaire intensity, the higher the threshold separating poor and rich

countries with regards to the sensitivity of their happiness self-evaluation to income

inequalities. In rich countries, income inequalities normally have a negative impact on

happiness, but the higher the level of billionaire intensity, the smaller this negative impact

is: Happiness = 3.7*** + 0.00008Ycap*** +1.7e-06GINIincome*** (22100 – Ycap +

+ 388* BillionaireIntensity***)

Table 2: Regression results of happiness indices on various determinants. Panel data

with controls for random and fixed effects (T-statistics add z- statistics in brackets)

Dependent variable – Happiness


22

*, **, *** -

denotes significance at 10, 5 and 1% respectively.

Model 2.4 shows the same relationship as model 2.3 (random effects), but with

controls for fixed-effects: Happiness = 5.0*** + 0.00007Ycap*** +1.9e-06GINIincome***

(15000 – Ycap*** + 312BillionaireIntensity)

This confirms the robustness of the estimates: the positive effect of income

inequalities on happiness for poor countries and the positive effect of billionaire intensity on

happiness for all countries is captured in space – cross-country comparisons, i.e., controlling

for random effects, as well as in time – time series for a particular country, i.e., controlling

for fixed effects. The only difference is the lower threshold: $15,000 compared to $22,000.

In model 2.5 – panel data without controls for fixed effects and non-linearity –

happiness indices are positively and significantly correlated to per-capita income, income

Model, parameters, N /

Variables

2.1,

Panel,

random

effects,

N=760

2.2,

Panel,

random

effects,

N=760

2.3,

Panel,

random

effects,

N=760

2.4,

Panel,

fixed

effects,

N=760

2.5,

Panel,

N=658

2.6,

Panel,

random

effects,

N=658

2.7,

Panel,

fixed

effects,

N=658

PPP GDP per capita, $,

Ycap

.00003***

(9.4)

.00009***

(5.6)

.00009***

(5.7)

.00007***

(3.9)

.00005***

(14.9)

.00003***

(8.2)

.00004***

(2.8)

Gini coefficient of

income distribution, %,

GINIincome

.02***

(3.2)

.04***

(4.6)

.04***

(4.6)

.02***

(2.7)

.03***

(4.7)

Gini coefficient of

wealth distribution, %,

GINIwealth

.02***

(4.2)

Billionaires’ wealth to

PPP GDP ratio, %,

Billionaires’Intensity

.03**

(2.4)

Murder rate (per

100,000 inhabitants),

MURDERrate

.18***

(5.4)

-.07***

(-5.6)

-.07***

(-5.4)

Interaction term

(Ycap * GINIincome)

-1.6e-06

***

(3.5)

-1.7e-06

***

(3.6)

-1.9e-06

***

(3.8)

-1.1e-06 **

(2.4)

Interaction term

(GINIincome *

*Billionaires’Intensity)

.0007

(1.6)

.0005

(1.2)

Interaction term

(GINIincome *

MURDERrate)

.002***

(6.2)

.001***

(5.2)

Constant 4.2***

(14.9)

3.7***

(11.7)

3.7***

(11.7)

5.0***

(13.8)

2.7***

(8.8)

5.2***

(8.2)

6.1***

(52.8)

R2 , % 41 42 43 29 47 41 17


23

inequality, wealth inequality, billionaire intensity, and murder rates. The interpretation could

be that inequality at all levels has a stronger positive impact on happiness – a positive impact

that outweighs the negative impact of a higher murder rate.

In model 2.6, an additional explanatory variable is introduced: an interaction term

between income inequalities and the murder rate. Counterintuitively, the impact of the

murder rate on happiness is positive for countries with high income inequalities – those with

Gini coefficients of income distribution of over 45%: Happiness = 5.2*** + 0.00003Ycap***

+ 0.0016MURDERS*** (GINIinc*** – 45)

This relationship most likely captures the patterns of poorer countries with high

income inequalities. For all rich countries, the Gini coefficient of income distribution is lower

than 45%, so countries with Gini coefficients of over 45% are all developing countries. Even

high murder rates in these countries do not prevent people from experiencing happiness

under high income inequalities, for example, in Latin America and Africa.

Finally, model 2.7 replicates the results with controls for fixed effects:

Happiness=6.1***+1.12e-06Ycap***(GINIinc***–38)+0.0014MURDERS***(GINIinc***–50)

This result is even stronger than with previous models: In countries with low income

inequalities – i.e., Gini coefficients below 38% – even income growth does not bring

happiness, and increase in murders undermines happiness.

In table 3, some results for the determinants of the murder rate and suicide rate are

reported. These indicators could be regarded as more objective measures of

wellbeing/happiness; i.e., if people are so unhappy with their lives and blame themselves,

they commit suicide; if they blame others, they commit murders. Reasons for suicides and

especially for murders may be different of course, but it is instructive to see the determinants

of both indicators anyway.

Models 3.1-3.2 suggest that the impact of income and wealth inequalities on murder

rates is positive, but the impact of wealth inequalities at the top (billionaire intensity) is

negative.

Model 3.3 shows that for countries with Gini coefficients of income inequality above

29% – most countries in the world – billionaire intensity has a negative impact on murder

rates: MURDERrate = 0.3GINinc*** + 0.06GINIwealth*** + 0.08Billionaires’Intensity***

(29 – GINIinc) – 8.4**

Suicide rates do not necessarily go together with high inequalities. Whereas one

might expect that people may feel less happy in countries with, and in times of, high levels


24

of inequality in income and wealth distribution, such that suicide rates in these countries and

these periods are higher, in fact, suicide rates do not exhibit any strong correlation with

income and wealth distribution.

Table 3: Regressions of murder rate and suicide rate on determinants

Dependent variable –

MURDERrate

Dependent variable –

SUICIDErate

Model, parameters, N /

Variables

3.1,

Panel,

N=876

3.2,

Panel,

random

effects,

N=876

3.3,

Panel,

random

effects,

N=876

3.4,

Panel,

N=197

3.5,

Panel,

random

effects,

N=197

3.6,

Panel,

fixed

effects,

N=197

PPP GDP per capita, $,

Ycap

-1.8e-06

**

(-2.5)

-3.5e-06

***

(-4.7)

-2.2e-06

(-1.3)

Gini coefficient of income

distribution, %, GINIincome

.8***

(15.4)

.3***

(5.1)

.3***

(5.3)

-.1***

(-2.7)

.06

(1.1)

.22***

(2.9)

Gini coefficient of wealth

distribution, %, GINIwealth

.2***

(5.6)

.06

(1.2)

-.2

(-1.6)

Billionaires’ wealth to PPP

GDP ratio, %,

Billionaires’Intensity

-.9***

(-6.5)

-.3***

(-2.7)

2.4***

(3.8)

Interaction term

(Ycap * GINIincome)

Interaction term

(GINIincome *

*Billionaires’Intensity)

-.08***

(-4.4)

Interaction term

(GINIincome *

MURDERrate)

Constant 34.9***

(9.6)

3.5

(1.5)

-8.4**

(-1.9)

16.4***

(8.3)

9.2***

(4.3)

15.4*

(2.0)

R2 , % 38 37 35 5 0 4

*, **, *** - denotes significance at 10, 5 and 1% respectively.


25

The best equations are reported in table 3: there is some indication that income

inequalities cause the suicide rate to fall (model 3.4), but once controls for random effects

are introduced, the relationship totally disappears (model 3.5 – R2 =0, all coefficients are

insignificant), whereas in the fixed effects model 3.6, income inequalities have a positive

impact on the suicide rate but wealth inequalities have a negative impact on the suicide rate.

Conclusions

To summarise, income and wealth inequalities do not always cause happiness levels to fall.

It is true that inequalities have an array of negative social consequences, well described in

the literature – from an increase in crime and mortality to a decline in educational attainment

and a proliferation of psychological disorders and obesity (Wilkinson and Pickett, 2010).

Besides, inequalities undermine social mobility and lead to the conservation of social

stratification: the higher the level of inequality, the higher the probability that one’s income

will closely resemble that of one’s parents – known as the ‘Great Gatsby’ curve. Hence the

social structure, and very often the political structure of society, becomes less flexible as

well.

The frequent claim that inequality promotes accumulation and growth does not get

much support from history. On the contrary, great economic inequality has always

been correlated with extreme concentration of political power, and that power has

always been used to widen the income gaps through rent-seeking and rent-keeping,

forces that demonstrably retard economic growth (Milanovic, Lindert, and Williamson,

2007).

As Joseph Stiglitz explains:

Widely unequal societies do not function efficiently, and their economies are neither

stable, nor sustainable in the long run…When the wealthiest use their political power

to benefit excessively the corporations they control, much needed revenues are

diverted into the pockets of a few instead of benefiting society at large… That higher

inequality is associated with lower growth – controlling for all other relevant factors –

have been verified by looking at the range of countries and looking over longer

periods of time” (Siglitz, 2012, p. 83, 117).


26

Latin American countries, writes Stiglitz, may show a glimpse of the future to other

states that are just stepping out on the road leading to growing inequalities.

“The experience of Latin American countries, the region of the world with the highest level

of inequality, foreshadows what lies ahead. Many of the countries were mired in civil conflict

for decades, suffered high levels of criminality and social instability. Social cohesion simply

did not exist” (Stiglitz, 2012, p. 84).

Developing countries with high levels of income inequality are more likely than others

to end up in a vicious circle: a bad equilibrium with poor quality institutions, low growth, low

levels of social mobility, and high social tensions. It may take a revolution to break this

vicious circle and to exit the bad equilibrium (Popov, 2014).

But some inequality is obviously not only tolerated by society, but is also

indispensable to obtaining the feeling of happiness. Wealth is increasingly not inherited, but

self-made, even in advanced countries (Freund and Oliver, 2016). The most rapid growth in

the number of self-made billionaires and in the growth in their wealth is in East Asia, but

such figures are also growing in Latin America and Sub-Saharan Africa.8 This may be the

reason why self-evaluations of happiness are often affected positively by income and wealth

inequalities, especially at the top of the distribution pyramid. In particular, there is strong

evidence that: (1) income and wealth inequalities in relatively poor countries – with personal

incomes below US$ 20,000-30,000 – affect happiness positively; and that (2) the relative

wealth of billionaires and millionaires – as a percentage of GDP – contributes to feelings of

happiness in poor and rich countries.

This paper contributes to the existing literature by specifying the non-linear impact of

inequality on happiness: income inequality raises happiness, rather than lowers it, in

relatively poor countries with per-capita income below $20,000-30,000, whereas in rich

countries lower inequality makes people feel better.

Inequality of wealth distribution – which is distinct from inequality of income because

wealth is a stock variable, whereas income is a flow – is positively linked to happiness in all

countries. Furthermore, wealth distribution at the very top of the pyramid – the billionaire

and millionaire wealth as a percentage of GDP – is one of the most important determinants

of happiness index scores in all countries. Wealth inequality is probably less irritating to

people than income inequality because wealth is associated with the past, i.e., it is always

inherited or acquired from the past, whereas income concerns the present, and current

injustices hurt more than past ones.


27

Consequently, there are two statistical portraits of a happy country: one has high

income per capita compared to the world average, a low level of income inequality, but high

wealth inequality and especially high wealth inequality at the very top – millionaire and

billionaire intensity. Very much like a typical Scandinavian country. The other type of a happy

country has lower levels of income, but high income and wealth inequalities, especially at

the top of the wealth pyramid. This is the Latin American and African model; e.g., Honduras,

Bolivia, Ecuador, Costa Rica, South Africa, and Zimbabwe. In both cases, wealth inequality

at the very top contributes to happiness more than inequality among other income groups.

The simplified picture of the happy society is that of 99% of the population having roughly

similar incomes at around the average level for these 99%, but where the remaining 1% are

millionaires and billionaires.

References

Alesina, A. and La Ferrara, E. (2001). Preferences for Redistribution in the Land of

Opportunities. NBER Working Paper, no. 8267.

Alesina, A., Di Tella, R., MacCulloch, R. (2001). Inequality and Happiness: Are Europeans

and Americans Different? NBER Working Paper, no. 8198.

Alesina, A. and Giuliano, P. (2009). Preferences for redistribution. NBER Working Paper

no. 14825.

Brockmann, H., Delhey, J., Welzel, C., and Yuan, H. (2008). The China Puzzle: Falling

Happiness in a Rising Economy. Journal of Happiness Studies, 10(4), pp. 387–405.

Chapman, J. (2018). Inequality and Poor Law Policy in Late Victorian England. Retrieved

from Http://eh.net/eha/wp-content/uploads/2018/06/Chapman.pdf.

Davies, J. B., Sandström, S., Shorrocks, A., and Wolff , E. N. (2007). Estimating the Level

and Distribution of Global Household Wealth. WIDER Research Paper, no. 77.

Retrieved from https://www.wider.unu.edu/publication/estimating-level-and-

distribution-global-household-wealth.

Easterlin, R. (2016). The science of happiness can trump GDP as a guide for policy. World

Economic Forum. Retrieved from https://www.weforum.org/agenda/2016/04/the-

science-of-happinesscan-trump-gdp-as-a-guide-for-policy.

Freund, C. and Oliver, S. (2016). The Origins of the Superrich: The Billionaire

Characteristics Database. The Petersen Institute for International Economics,

Working Paper no 16-1. Retrieved from https://piie.com/publications/wp/wp16-1.pdf.


28

Credit Suisse. (2018). Global Wealth Report; Global Wealth Databook 2018. Retrieved

from https://www.credit-suisse.com/corporate/en/research/research-institute/global-

wealth-report.html.

Helliwell, J., Layard, R., and Sachs, J. (2018). World Happiness Report 2018. New York:

Sustainable Development Solutions Network.

Sundaram, J. K. and Popov, V. (2015). Income Inequalities in Perspective. Development,

58(2–3), pp. 196–205.

Kuznets, S. (1955). Economic Growth and Income Inequality. American Economic Review,

45, pp. 1–28.

Marsh, H. W. and Parker, J. W. (1984). Determinants of student self-concept: Is it better to

be a relatively large fish in a small pond even if you don’t learn to swim as well?

Journal of Personality and Social Psychology, 47(1), pp. 213–231.

Milanovic, B. (2008). Where in the world are you? Assessing the importance of

circumstance and effort in a world of different mean country incomes and (almost)

no migration. The World Bank, Working Paper S4493.

Piketty, T., Saez, E., and Zucman, G. (2016). Distributional National Accounts: Methods

and Estimates for the United States. NBER Working Paper, no. 22945. Retrieved

from https://www.nber.org/papers/w22945.

Popov, V. (2014). Mixed Fortunes: An Economic History of China, Russia and the West.

Oxford: Oxford University Press.

Popov, V. (2018a). Paradoxes of happiness: Why do people feel more comfortable with

high levels of inequality and high murder rates? DOC Research Institute. Retrieved

from https://doc-research.org/2018/06/paradoxes-happiness/.

Popov, V. (2018b). Why do some countries have more billionaires than others? Explaining

variations in the billionaire intensity of GDP. DOC Research Institute. Retrieved

from https://pages.nes.ru/vpopov/documents/Billionaires_Web-upload.pdf.

Sachs, J. D. (2018). America’s Health Crisis and the Easterlin Paradox. In J. Helliwell, R.

Layard, and J. Sachs (Eds.), (2018). World Happiness Report 2018 (Chapter 7, pp.

146–159). New York: Sustainable Development Solutions Network.

Stiglitz, J. (2012). The price of inequality: How today's divided society endangers our

future. New York: W.W. Norton & Co.

UNODC (2019). United Nations Office on Drugs and Crime.

WDI (World Development Indicators) database (2018).


29

WHO (2018). World Health Organization. Causes of death statistics. Retrieved from

Http://apps.who.int/gho/data/node.main.GHECOD?lang=en.

Wilkinson, R., Pickett, K. (2010). The Spirit Level: Why Greater Equality makes Societies

Stronger. New York: Bloomsbury Press.

World Database on Happiness (2019). Retrieved from

https://worlddatabaseofhappiness.eur.nl/.

1 The happiness index in this paper is taken from the World Database on Happiness. This is a self-evaluation of personal happiness on a scale from 0 to 10, derived from surveys. 2 This model predicts that in a majority-rule voting system, the outcome selected at the polls will be the one preferred by the median voter, i.e., the voter separating one half of the electorate from the other half, if all voters are ranked according to their preferences. 3 HappinessIndex = 5.8*** – 9.7e-07 Ycap + 0.007MURDERrate* – 0.01SUICIDErate

N=323 (Panel data for 2000-18 for over 200 countries, some observations are missing), R2=0.014, robust estimates, no control for fixed and random effects

HappinessIndex = 5.7*** + 0.005MURDERrate – 0.02SUICIDErate*

Random effects regression, R2 (between) = 0.02, N=328.

HappinessIndex = 5.7*** – 1.4e-08 Ycap + 0.005MURDERrate – 0.03SUICIDErate**

Random effects regression, R2 (between) = 0.02, N=323.

Here and later the following notations are used: *, **, *** – denotes significance at 10, 5 and 1% level respectively. 4 (50:10) / (50:90) = 9 5 This is how the estimation procedure is explained in the GWR: “We exploit the fact that the top tail of wealth distribution is usually well approximated by the Pareto distribution, which produces a straight-line graph when the logarithm of the number of persons above wealth level w is plotted against the logarithm of w. Our data yield a close fit to the Pareto distribution in the wealth range from USD 250,000 to USD 5 million. Above USD 5 million the relationship begins to break down, and the correspondence weakens further above USD 50 million, as expected given the limitations of the data sources. However, it still seems reasonable to use a fitted Pareto line to estimate the number of individuals in the highest echelons of the wealth distribution. To determine the precise features of the top wealth tail, we rely heavily on the rich list data provided by Forbes and other sources. We make particular use of the number of billionaires reported by Forbes, since the data are available for many years and are broadly comparable across countries. We recognize that rich list data have limitations. The valuations of individual wealth holdings are dominated by financial assets, especially equity holdings in public companies traded in international markets. For practical reasons, less attention is given to nonfinancial assets apart from major real estate holdings and trophy assets, such as expensive yachts. Even less is known – and hence recorded – about personal debts. Some people cooperate enthusiastically with those compiling the lists; others jealously guard their privacy. There are also different country listings for nationals and residents, which is especially evident for India, for instance. The true legal ownership within families – as opposed to nominal ownership or control – adds further complications. Assigning the wealth recorded for Bill Gates, for example, to all family members might well result in several billionaire holdings, so the number of billionaires would increase in this instance. In other cases, reassigning the family wealth would reduce all the individual holdings below the billionaire threshold. For all these reasons, rich list data should be treated with caution. At the same time, the broad patterns and trends are informative, and they provide the best available source of information at the apex of global wealth distribution” (GWR, 2018, p. 111). 6 External causes of death include murders, suicides, accidents, and ‘unidentified’. 7 Billionare’sIntensity = 2.5***– 0,04GINIincome***

Robust estimate, R2 = 0.03, N=1031 (over 200 countries and 18 years, but some observations are missing).

Billionare’sIntensity = 1.9*** – 1.7e-07 Ycap – 0,03GINIincome**

Random effect regression, R2 (between) = 0.05, N=1031 (over 200 countries and 18 years, but some observations are

missing).

The following notations are used: *, **, *** – denotes significance at 10, 5 and 1% level respectively.


30

8 The Middle East and North Africa is the only region where the share of inherited wealth is growing and the wealth share of company founders is falling.

Billionaires, millionaires, inequality, and happiness · Billionaires, millionaires, inequality,...

Documents

Transcript of Billionaires, millionaires, inequality, and happiness · Billionaires, millionaires, inequality,...