Trends in global income inequality and their political implications

Trends in global income inequality and their political implications LIS Center; Graduate School City University of New York May 2015

A. NaEonal inequaliees mostly increased

Ginis in the late 1980s and around now 1985-90 A)er 2008 Change Average Gini 36.3 38.8 +2.5 Pop- weighted Gini GDP- weighted Gini Countries with higher Ginis 33.9 37.3 +3.4 32.2 36.4 +4.2 32.0 36.2 +4.5 Countries with lower Ginis From final- complete3.dta and key_variables_calcul2.do (lines 2 and 3; rest from AlltheGinis) 42.8 39.5-3.3

Ginis in the late 1980s and around now Gini after 2008 20 30 40 50 60 70 CRI CHN ECUDOM BOL SGP NGA USA MYS PER SLV URY CIV ARG UGA GEO MKD ISR KOR MRT TUR RUS IRN IDN PHL THA VEN GBR KGZ LVA BGR PRT JOR LKA ITA IND CAN MDA LTUPOL ESP FRA GRC ROU MLI EST BGD JPN DEU TWN AZE HRV TJK IRL PAK AUS FIN AUT BEL NLD HUNKAZ ARM CZE SVKUKR BLR SWE DNK NOR SVN COL MEX GTM HND PAN CHL BRA 20 30 40 50 60 Gini between 1985 and 1990 twoway (sca9er bbb aaa if year==2000, mlabel(contcod) msize(vlarge)) (funcfon y=x, range(20 60) legend(off) xftle(gini between 1985 and 1990) yftle(gini a)er 2008)) using allginis.dta

Ginis in 1988 and 2008 (populaeon- weighted countries) Gini in 2008 20 30 40 50 60 CHN-U RUS CHN-R IND-R USA NGA IND-U MEX BRA 20 30 40 50 60 Gini in 1988 From twenty_years/ key_variables_calcul3.do

Convergence of countries Ginis: an empirical observaeon without theoreecal explanaeon -20-10 0 10 20 BGR HUN POL CZE ECU GTM ARG CHN CHL GBR USA NZL SYC JAM HKG DOM SGP IND ISR VEN PAN PRI COL BELTWN IDN IRN CAN FJI AUS GRC KOR LKA SDN ZMB THA CRI SLV ESP BOL SWENLDPAK IRL MEX JPN BGD PRT BHS DEU FIN NOR ITA DNK EGY BRB MYS PHL TUN FRA TZA TTO TUR SLE NPL 20 30 40 50 60 average country Giniall before 1980 BRA PER GAB HND twoway (sca9er change_gini gini_pre1980 if nvals==1, mlabel(contcod)) (lfit change_gini gini_pre1980, yline(0, lpa9ern(dash)) yftle(change in Gini a)er 1980) legend(off)) Using Allthe Ginis.dta

Market, gross and disposable income Ginis in the US and Germany USA Germany.25.3.35.4.45.5 1970 1980 1990 2000 2010 year.25.3.35.4.45.5 1970 1980 1990 2000 2010 year Define_variables.do using data_voter_checked.dta

Issues raised by growing naeonal inequaliees Social separaesm of the rich Hollowing out of the middle classes Inequality as one of the causes of the global financial crisis PercepFon of inequality outstrips real increase because of globalizaeon, role of social media and poliecal (crony) capitalism (example of Egypt) Hidden assets of the rich

Some long- term examples set in the Kuznets framework

50.0 Inequality (Gini) in the USA 1929-2009 (gross income across households) 48.0 46.0 44.0 42.0 40.0 38.0 1929 1939 1949 1959 1969 1979 1989 1999 2009 From ydisrt/us_and_uk.xls

Kuznets and Pikedy frames 70 Ginis for England/UK and the United States in a very long run 60 50 USA 40 30 England/UK 20 10 0 1600 1650 1700 1750 1800 1850 1900 1950 2000 2050 From uk_and_usa.xls 11

Contemporary examples of Brazil and China: moving on the descending poreon of the Kuznets curve Brazil 1960-2010 China, 1967-2012 Gini 40 50 60 Gini 40 50 60 7.5 8 8.5 9 9.5 ln GDP per capita updated Giniall Fitted values 6 7 8 9 10 ln GDP per capita twoway (scatter Giniall lngdpppp if contcod=="bra", connect(l) ylabel(40(10)60) xtitle(2000 6000 12000) ytitle(gini) xtitle(ln GDP per capita)) (qfit Giniall lngdpppp if contcod=="bra", lwidth(thick)) From gdppppreg4.dta twoway (scatter Giniall lngdpppp if contcod=="chn" & year>1960, connect(l) ylabel(40(10)60) xtitle(2000 6000 12000) ytitle(gini) xtitle(ln GDP per capita)) (qfit Giniall lngdpppp if contcod=="chn" & year>1960, lwidth(thick)) From gdppppreg4.dta 12

B. Between naeonal inequaliees remained very high even if decreasing

DistribuEon of people by income of the country where they live: empeness in the middle (year 2013; 2011 PPPs) Percent 0 10 20 30 India, Indonesia China Brazil, Mexico, Russia W.Europe, Japan USA 0 10000 20000 30000 40000 50000 GDP per capita in 2005 PPP From defines.do in interyd

Different countries and income classes in global income distribueon in 2008 percentile of world income distribution 1 10 20 30 40 50 60 70 80 90 100 Russia USA China India Brazil 1 20 40 60 80 100 country percentile From calcu08.dta

1 10 20 30 40 50 60 70 80 90 100 Denmark Mozambique Mali Tanzania Uganda 1 5 10 15 20 country ventile

14 Countries with more than 1% of their populafon in top global percenfle (above $PPP 72,000 per capita in 2008 prices) 12 12 10 9 9 8 6 5 6 7 7 4 2 2 2 2 2 2 2 3 3 0 CYP DEU IRL KOR NLD TWN FRA NOR GBR JPN CAN LUX CHE SGP USA From summary_data.xls

C. Global inequality is the product of within- and between- county inequaliees How did it change in the last 60 years?

EssenEally, global inequality is determined by three forces What happens to within- country income distribueons? Is there a catching up of poor countries? Are mean incomes of populous & large countries (China, India) growing faster or slower that the rich world?

Global and internaeonal inequality aper World War II Gini coefficient.45.55.65.75 Concept 2 Concept 1 Concept 3 Within- naeonal inequaliees 1950 1960 1970 1980 1990 2000 2010 year Defines.do using gdppppreg5.dta Concept2: 1960-1980 from Bourguignon & Morrisson

Concept 2 inequality with 2011 PPPs and without China and India.45.5.55.6.65 all countries Without China Without India and China 47 1940 1960 1980 2000 2020 year Defines.do using gdppppreg5.dta

Non- triviality of the omided countries (Maddison vs. WDI) PopulaEon coverage 1988 1993 1998 2002 2005 2008 2011 Africa 48 76 67 77 78 78 70 Asia 93 95 94 96 94 98 96 E.Europe 99 95 100 97 93 92 87 LAC 87 92 93 96 96 97 97 WENAO 92 95 97 99 99 97 96 World 87 92 92 94 93 94 92

Three important technical issues in the measurement of global inequality The ever- changing PPPs in parecular for populous countries like China and India The increasing discrepancy between GDP per capita and HS means, or more importantly consumpeon per capita and HS means Inadequate coverage of top 1% (related also to the previous point)

The issue of PPPs

The effect of the new PPPs on countries GDP per capita (compared to the US level) -50 0 50 100 150 SAU ZMB SDN JOR GHA IDN MNG SUR OMN KWT PAK EGY NPLBGD FJI AZE KAZ QAT YEM CIV LAO CPV DZA THA MDG LKA MAC VNMPHLGTM NER BRN MLI MAR HTITCD COG VENRUS GNQ MYS ARE TGO KEN MRT IND MDV LSO BDI SLE UGA KGZ NGAMDA AGONAM BRA GIN CMR SWZ CHN KHM BTNUKR ETH BLR TJK NIC BOL TUNMKD GNB RWA BFA BEN SEN GEO PRY MNE ARM BIH BGR TUR LVA CAF MWI HND SLV BLZECU DOM PER HUN SGP COL MEX URY CHL TTO SRB ZAF LTU EST MUS HRV POL TZA JAM CRIGAB SVK MLT ITA DNKCHE NOR LUX NZL PAN PRT GRCESP FRA FIN TWN BEL DEU SWE IRLUSA MOZ DJI ALB CZE SVN ISR ISL AUT AUS CAN NLD JPN HKG LBR KOR GBR GMB BWA CYP COM gdppc in 2011ppp BHS 50000 100000 150000 C:\Branko\worldyd\ppp\2011_icp\define

Country The effect of new PPPs GDP per capita increase (in %) GDP per capita increase populafon- weighted (in %) Indonesia 90 - - - Pakistan 66 - - - Russia 35 - - - India 26 - - - China 17 - - - Africa 23 32 Asia 48 33 LaEn America 13 17 Eastern Europe 16 24 WENAO 3 2

Global income inequality using nominal dollars Gini coefficient.55.6.65.7.75.8.85 Concept 1 Concept 2 Concept 3 1970 1980 1990 2000 2010 Year 63 From two_concepts_exrate.do using Global_new5.dta

The gap between naeonal accounts and household surveys

Both the level and change: Use of GDP per capita gives a lower lever and a faster decrease of global inequality Gini.45.5.55.6.65 GDPs pc countries in HS sample HS means--countries in HS sample usual Concept 2 1990 1995 2000 2005 2010 2015 year Defines.do based on gdppppreg5.dta

How global inequality changes with different definieons of income 72 71 70 69 68 67 66 65 64 63 62 Step 1 Global inequality Step 2 GDP ppp ConsumpEon Survey mean

Step 1 driven by low consumpeon shares in China and India (although on an unweighted base C/GDP decreases with GDP) C/GDP from national accounts in year 2008.2.4.6.8 1 1.2 India China USA 1000 10000 50000 GDP per capita in ppp twoway sca9er cons_gdp gdpppp if group==1 & cons_gdp<1.4 [w=totpop], xscale(log) xftle(gdp per capita in ppp) xlabel(1000 10000 50000) yftle(share of consumpfon in GDP) Ftle(C/GDP from nafonal accounts in year 2008) using final08,dta

Step 2. No clear (weighted) relaeonship between survey capture and NA consumpeon survey mean/consumption from national account in year 2008.2.4.6.8 1 1.2 India China USA 1000 10000 50000 GDP per capita in ppp twoway sca9er scale2 gdpppp if group==1 & scale2<1.5 [w=totpop], xscale(log) xftle(gdp per capita in ppp) xlabel(1000 10000 50000) yftle(survey mean over NA consumpfon) Ftle(survey mean/consumpfon from nafonal account in year 2008)

The issue of top underesemaeon

Rising HS/NA gap and top underesemaeon If these two problems are really just one & the same problem. Assign the enere posieve (NA consumpeon HS mean) gap to naeonal top deciles Use Pareto interpolaeon to elongate the distribueon No a priori guarantee that global Gini will increase

Gini: accouneng for missing top incomes 1988 1993 1998 2003 2008 Surveys only NAC instead of survey mean NAC with Pareto NAC with top- heavy Pareto 72.5 71.8 71.9 71.9 69.6 71.5 70.5 70.6 70.7 67.6 71.8 70.8 71.0 71.1 68.0 76.3 76.1 77.2 78.1 75.9

The results of various adjustments Replacing HS survey mean with private consumpeon from NA reduces Gini by 1 to 2 points ElongaEng such a distribueon (that is, without changing the consumpeon mean) adds less than ½ Gini point But doing the top- heavy adjustment (NA- HS gap ascribed to top 10% only) adds between 5 and 7 Gini points It also almost eliminates the decrease in global Gini between 1988 and 2008

How Global Gini in 2008 changes with different adjustments 10 Increase in global Gini with each marginal adjustment 8 6 4 2 Allocate the gap proporeonally along each naeonal income distribueons Allocate the gap proporeonately and add a Pareto elongaeon Allocate the gap to top 10% and add Pareto elongaeon 0-2 - 4

With full adjustment (allocaeon to the top 10% + Pareto) Gini decline almost fully disappears 80 78 76 74 72 Top- heavy allocaeon of the gap + Pareto adjustment Survey data only 70 68 66 64 1988 1993 1998 2003 2008

D. How has the world changed between the fall of the Berlin Wall and the Great Recession [based on joint work with Christoph Lakner]

Real PPP income change (in percent) Real income growth at various perceneles of global income distribueon, 1988-2008 (in 2005 PPPs) 80 70 60 50 40 30 20 10 $PPP2 X China s middle class $PPP4.5 $PPP12 $PPP 110 X US lower middle class 0 0 10 20 30 40 50 60 70 80 90 100 PercenFle of global income distribufon From twenty_years\final\summary_data EsEmated at mean- over- mean

Real PPP income change (in percent) 90 80 70 60 50 40 30 20 10 0-10 Real income gains (in $PPP) at different percenfle of global income distribufon 1988-2008 World Without China 0 10 20 30 40 50 60 70 80 90 100-20 PercenFle of global income distribufon

Quasi non- anonymous GIC: Average growth rate 1988-2008 for different perceneles of the 1988 global income distribueon

Growth incidence curve (1988-2008) esemated at perceneles of the income distribueon 0 20 40 60 80 mean growth 2 10 20 30 40 50 60 70 80 90 95 100 percentile of global income distribution Using my_graphs.do Mean- on- mean

DistribuFon of global absolute gains in income, 1988-2008 (anonymous) 30.0 25.0 25 DistribuFon (in percent) of gain 20.0 15.0 10.0 5.0 0.0 From summary_data.xls 19 16 8 5 4 4 0 0 0 1 1 1 1 1 1 2 2 2 3 3 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 99 100 venfle/percenfle of global income distribufon

5000 Annual per capita a)er- tax income in internafonal dollars US 2nd decile Chinese 8th urban decile 500 1988 1993 1998 2003 2008 2011 From summary_data.xls

Global income distribueons in 1988 and 2008 density 0.2.4.6.8 1 1988 2008 Emerging global middle class between $3 and $16 300 1000 3000 6000 10000 30000 50000 100000 log of annual PPP real income twoway (kdensity logrrinc [w=pop] if logrrinc>2 & bin_year==2008 & keep==1 & mysample==1) (kdensity logrrinc [w=pop] if logrrinc>2 & bin_year==1988 & keep==1 & mysample==1, legend(off) xetle(log of annual PPP real income) yetle(density) text(0.95 2.5 "1988") text(0.85 3 "2008")) Or using adding_xlabel.do; always using final_complete7.dta

Increasing gains for the rich with a widening urban- rural gap Urban and rural China Urban and rural Indonesia 200 250 300 350 400 450 urban rural 170 180 190 200 210 220 urban rural 1 2 3 4 5 6 7 8 9 10 decile 1 2 3 4 5 6 7 8 9 10 decile From key_variables_calcul2.do

E. Issues of jusece and poliecs 1. CiEzenship rent 2. MigraEon 3. Hollowing out of the middle classes

Global inequality of opportunity Regressing (log) average incomes of 118 countries perceneles (11,800 data points) against country dummies explains 77% of variability of income perceneles Where you live is the most important determinant of your income; for 97% of people in the world: birth=ciezenship. CiEzenship rent.

Is ciezenship a rent? If most of our income is determined by ciezenship, then there is lidle equality of opportunity globally and ciezenship is a rent (unrelated to individual desert, effort) Key issue: Is global equality of opportunity something that we ought to be concerned or not? Does naeonal self- determinaeon dispenses with the need to worry about GEO?

The logic of the argument CiEzenship is a morally- arbitrary circumstance, independent of individual effort It can be regarded as a rent (shared by all members of a community) Are ciezenship rents globally acceptable or not? PoliEcal philosophy arguments pro (social contract; staest theory; self- determinaeon) and contra (cosmopolitan approach)

The Rawlsian world For Rawls, global opemum distribueon of income is simply a sum of naeonal opemal income distribueons Why Rawlsian world will remain unequal?

Global inequality in Real World, Rawlsian World, Convergence World and Shangri- La World (Theil 0; year 2008) Individual incomes within country All equal Mean country incomes Different (as now) All equal 0 30 (all mean incomes equalized; all country Ginis as now) Different (as now) 68 (all country Theils=0; all mean incomes as now) 98

Conclusion Working on equalizaeon of within- naeonal inequaliees will not be sufficient to significantly reduce global inequality Faster growth of poorer countries is key and also

MigraEon: a different way to reduce global inequality and ciezenship rent A new view of development: Development is increased income for poor people regardless of where they are, in their countries of birth or elsewhere MigraEon and LDC growth thus become the two equivalent instruments for development

A migrant point of view: trade- off between country s mean income and its inequality 14 How much is one Gini point change worth in terms of mean country income? 12 10 Percent of income 8 6 4 Decrease in Gini Increase in Gini 2 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Ventile From interyd..\venel_vs_country.xls

PoliEcal issue: Global vs. naeonal level Our income and employment is increasingly determined by global forces But poliecal decision- making sell takes place at the level of the naeon- state If stagnaeon of income of rich countries middle classes conenues, will they conenue to support globalizaeon? Two dangers: populism and plutocracy To avert both, need for within- naeonal redistribueons: those who lose have to be helped

Final conclusion To reduce global inequality: fast growth of poor countries + migraeon To preserve good aspects of globalizaeon: redistribueon within rich countries

AddiEonal slides

H. Global inequality over the long- run of history

Global income inequality, 1820-2008 (Source: Bourguignon- Morrisson and Milanovic; 1990 PPPs ) 0 20 40 60 80 100 Theil Gini 1820 1860 1900 1940 1980 2020 year twoway (sca9er Gini year, c(l) xlabel(1820(40)2020) ylabel(0(20)100) msize(vlarge) clwidth(thick)) (sca9er Theil year, c(l) msize(large) legend(off) text(90 2010 "Theil") text(70 2010 "Gini"))

Shares of global income received by top 10% and bodom 60% of world populaeon 70 60 Top 10% (L- M data) Percentage share of global income 50 40 30 20 Top 10% (B- M data) Bodom 60% (B- M data) 10 Bodom 60% (L- M data) 0 1800 1850 1900 1950 2000 2050 Year

A non- Marxist world Over the long run, decreasing importance of within- country inequaliees despite some reversal in the last quarter century Increasing importance of between- country inequaliees (but with some hopeful signs in the last five years, before the current crisis), Global division between countries more than between classes

ComposiEon of global inequality changed: from being mostly due to class (within- naeonal), today it is mostly due to locaeon (where people live) 100 80 Theil 0 index (mean log deviation) 60 40 20 Location Class Location Class 0 1870 2008 Based on Bourguignon- Morrisson (2002), Maddison data, and Milanovic (2005) From thepast.xls

Very high but decreasing importance of locaeon in global inequality 90 Share of the between component in global Theil (0) Between component, in percent 80 70 60 50 40 30 20 B- M data L- M data 10 0 1800 1850 1900 Year 1950 2000 2050 From thepast.xls under c:\history