European Historical Economics Society EHES WORKING PAPERS IN ECONOMIC HISTORY | NO. 131 Well-being Inequality in the Long Run Leandro Prados de la Escosura Universidad Carlos III, Groningen, and CEPR MAY 2018
European
Historical
Economics
Society
EHES WORKING PAPERS IN ECONOMIC HISTORY | NO. 131
Well-being Inequality in the Long Run
Leandro Prados de la Escosura
Universidad Carlos III, Groningen, and CEPR
MAY 2018
EHES Working Paper | No. 131|May 2018
Well-being Inequality in the Long Run1
Leandro Prados de la Escosura*
Universidad Carlos III, Groningen, and CEPR
Abstract
This paper provides a long-run view of well-being inequality at world scale based on a new historical
dataset. Trends in social dimensions alter the view on inequality derived from per capita GDP. While in
terms of income, inequality increased until the third quarter of the twentieth century; in terms of well-
being, inequality fell steadily since World War I. The spread of mass primary education and the health
transitions were its main drivers. The gap between the West and the Rest explains only partially the
evolution of well-being inequality, as the dispersion within the developing regions has increasingly
determined its evolution.
JEL classification: I00, N30, O15, O50
Keywords: Well-being, Inequality, Life Expectancy, Health Transition, Education, per capita GDP.
1 I gratefully acknowledge comments by participants at the 4th World Bank-Banco de España Policy Conference (Madrid), Economic History
Society Conference (Cambridge), the GGDC 25th Anniversary Conference (Groningen), the World Congress of Cliometrics (Strasbourg), and
seminars at Bar-Ilan (Tel Aviv), NYUAD (Abu Dhabi), Research School of Economics, ANU (Canberra), and Higher School of Economics
(Moscow). A research grant from Fundación Rafael del Pino is acknowledged.
* Universidad Carlos III, Department of Social Sciences, Madrid, 135, 28903 Getafe (Madrid), Spain [email protected].
Notice
The material presented in the EHES Working Paper Series is property of the author(s) and should be quoted as such.
The views expressed in this Paper are those of the author(s) and do not necessarily represent the views of the EHES or
its members
3
Introduction
Inthelastoneandahalfcenturies,substantialgainsacrosstheboardare
observedformainwell-beingdimensions(includingpercapitaGDP,health,education,
politicalvoice,civilliberties,andpersonalsecurity)(Maddison,2006;Bourguignonand
Morrisson,2002;MorrissonandMurtin,2009;PradosdelaEscosura,2015;Pinker,
2018).Howhavethesegainsbeendistributed?Doinequalitytrendsinwell-being
dimensionsconcur?DidthegapbetweentheWestandtheRestoftheworldexplain
well-beinginequality?
Thispaperapproacheslong-runwell-beinginequalityfromamultidimensional
perspectiveinspiredbythecapabilitiesapproach.2Thecapabilitiesapproachmakes
well-beingdependentonacombinationoffunctionings(orachievements)and
capabilities(theabilitytochooseamongalternativebundlesoffunctionings)3.Thus,
well-beingdifferencesacrosscountrieswillbeassessedonabroadbasisthatincludes
notonlytheeconomicdimension(realpercapitaincome),butalsohealth(life
expectancyatbirth)andeducation(literacyandgrossenrolmentratesandyearsof
schooling)dimensions.Differencesinhumandevelopment,'aprocessofenlarging
people’schoices'(UNDP,1990),whichencompassesenjoyingahealthylife,acquiring
knowledge,andachievingadecentstandardofliving,willbealsoaddressed.
Thedatabasecomprisesalargegroupofcountriesrangingbetween96and164
andrepresentingover90percentoftheworldpopulation.Thetimespanconsidered
coversfromthefirstglobalization,thebeginningsofmassprimaryeducation,andthe
eveoftheepidemiologicaltransitiontothepresent,underanotherphaseof
globalization,withtertiaryeducationspreadingatworldscale,andanewhealth
transitionunderway.
Acaveatisneeded.Inthepresentstateoftheart,itisnotpossibletoderive
measuresofwithin-countryinequalityfortheselectedwell-beingdimensionsover
2Alternativesarethewelfareeconomicsapproachwhichvaluesvariousdimensionsofqualityoflife,includinghealth,education,environment,etc.inmonetaryterms(NordhausandTobin,1972;Beckeretal.,2005;JonesandKlenow,2016;GallardoAlbarrán,2017)andtheSubjectiveWell-Being(SWB)approach,whichplaceslifesatisfactionatitscentre(Easterlin,1974,KahnemanandDeaton,2010;VeenhovenandVengust,2013).3Thispaperisonlyinspiredonthecapabilitiesapproachbecause,sofar,Ihavefocusedonachievementsonly.SeeIvanovandPeleah(2010).
4
suchalargesampleandlongtimespan.Thediscussionwilladdress,then,inter-
nationalinequality,inwhichnationalaveragesareemployed.
Thepaperisorganizedasfollows.InsectionIIanoverviewoftheempirical
literatureonmultidimensionalinternationalinequalityprovideshypothesestobe
exploredfromalongrunperspective.SectionIIIpresentsnewwell-beingindicesoflife
expectancy,literacyandschoolenrolmentrates,andyearsofschooling,andhuman
development.InsectionIV,long-runtrendsininequalityareprovidedforeachwell-
beingindicatorand,then,abreakdownofinequalityintothedispersionwithinthe
WestandtheRestandthegapbetweenthem.Thelastsectionrecapsandraises
questionsforfurtherresearch.
Theresultschallengetheviewonlongruninequalityderivedfromrealper
capitaGDP.InequalityinsocialdimensionsdeclinedafterWorldWarI,unlikeincome
inequalitythatincreaseduntilthelatetwentiethcenturyandonly,then,declined.The
spreadofmassprimaryeducationandthehealthtransitionsdrovethedeclineofwell-
beinginequality.Moreover,itsevolutionisonlypartiallyexplainedbytheWest-Rest
gapasthedispersionwithindevelopingregionsprogressivelydefineditstrends.
DebatingWell-beingInequality
Earlierquantitativeassessmentsofinternationalinequalitywerecarriedouton
thebasisofpercapitaGDP,focusingalmostexclusivelyonthelatetwentiethcentury.
Along-termdeteriorationinworlddistributionofincome,ledbya wideninggap
betweendevelopedanddevelopingcountries,wastheprevailingconsensusuptothe
1980s(Theil1979,1989).AlbertBerry,FrançoisBourguignon,andChristianMorrisson
(1983)challengedthisviewbypointingoutthatlargecountrieswerethemain
determinantsoftheexhibitedtrends.4Later,BrankoMilanovic(2005,2016)showed
thatinternationalpopulation-weightedinequality(‘Inequality2’,inhistypology)fell
sincethemid-twentiethcenturywhileunweightedinequality(‘Inequality1’)
experiencedasustainedincreaseuntil2000and,then,declined.
4ThedeterminantroleplayedbyChinaandIndiaintheinternationalincomedistributionisarecurrentfeatureinlaterstudies,cf.Firebaugh(1999),BourguignonandMorrisson(2002),Milanovic(2005),andSala-i-Martin(2006).
5
Inthelastdecades,asdataonhouseholdsurveyshasbecomewidelyavailable,
researchshiftedthefocusto‘global’economicinequality,thatis,incomedistribution
amongindividuals,notjustacrosscountries’averages.Theresultsfromthenew
approachinitiallysupportedtheviewofasubstantialincreaseinglobalinequalityas
wideninginter-countryincomedifferentialsmorethanoffsetthedeclineinwithin-
countryinequality(KorzeniewiczandMoran,1997).Theconsensuswasbrokenwhen
PaulSchultz(1998)showedthatinequalityhadfallensincethemid-1970s,asthe
contractionininequalityacrosscountriescancelledanyincreasesinwithin-country
inequality.Schultz’sfindingsledtoanew,lesspessimisticconsensusthatchallenged
theviewofawideninggapbetweentheworldrichandpoorinthelatetwentieth
century.GlennFirebaugh(1999)alsonoticedaremarkablestabilityinworldincome
distributionbetween1960and1989sincethedivergenceinincomegrowthfavourable
torichcountrieswasoffsetbythefasterpopulationgrowthinpoorcountries.Forthe
post-1980era,MilanovicandJohnRoemer(2016)indicatethatthelevelofglobal
inequality(‘Inequality3’)remainedstableandhigh,andonlydeclinedsincetheearly
2000s.5
Longruninequalityhasreceivedlittlequantitativeattentionduetodata
constraints.BourguignonandMorrisson(2002),onthebasisof33“trans-national”
units,concludedthatworldinequalitywasmuchhigherin1992thanin1820.This
resultedfromariseininequalitybetweentheearlynineteenthcenturyandmid-
twentiethcenturythattendedtostabilizeduringthesecondhalfofthecentury.The
mainelementbehindlongrunworldincomeinequalitywasthedisparityacross
countries.Nonetheless,within-countryincomedistributiondominatedworld
inequalityduringthenineteenthcentury,whileinthetwentiethcenturycross-country
incomedistributionprevailed.MorerefinedestimatesbyJan-LuitenvanZandenetal.
(2014)tendedtoconfirmBourguignonandMorrisson’sfindings.Drivenbybetween-
countryinequality,thedispersionofglobalincomedistributionincreasedoverthelong
run,mostlyupto1950,stabilisingthereafter,andexperiencingamoderaterisefrom
1980onwards.
5DowrickandAkmal(2005)found,however,thatinequalityincreasedslightlybetween1980and1993.Liberati(2015)confirmedMilanovic’sfindingandpointedoutamoderatedeclineinglobalinequalitysincethebeginningofthetwentieth-firstcentury.
6
Percapitaincomeisjustonewell-beingdimensionandsocialscientistshave
alsobeenlookingatotherssuchashealthandeducation.MorrissonandMurtin(2013)
providealongrunviewofglobaleducationinequalityonthebasisofaverageyearsof
schooling6,findingalong-termreduction,mostlyattributabletothediffusionof
literacy.GoeslingandBaker(2008)observedthedecliningdispersionofschooling
yearssince1980andattributedittotheglobalizationofprimaryeducation.
ThelongrunevolutionofhealthinequalityhasbeenaddressedbyBourguignon
andMorrisson(2002)whoestimatedcross-countryinequalityforlifeexpectancyat
birthatscatteredbenchmarks,findingasustainedincreaseininequalitybetween1820
and1910,thatstabilizedupto1929,and,then,declinedsharplydownto1970,
remainingunaltereduntil1990.Anupdateoftheestimatesshowsafurtherdeclinein
the1990s(MorrissonandFabriceMurtin,2005).Thus,thelatetwentiethcentury
levelsofinequalityweresimilartothoseoftheearlynineteenthcentury.7Brian
GoeslingandFirebaugh(2004)foundadeclineinthe1980sthatwasrevertedduring
the1990s,andtheylargelyattributedtothedifferentpaceatwhichlifeexpectancy
evolvedacrosscountries,particularlyinSub-SaharanAfrica.RatiRam(2006)concurred
pointingoutthecontrastbetweenthepre-andpost-1990periodsandattributingthe
divergenceinthe1990stotheroleofHIV/AIDS.GoeslingandDavidBaker(2008)
stressedtheunevendiffusionofhealthknowledge,practice,andtechnologyacross
countriesthat,asRyanEdwards(2011)observes,translatedintosteady,oreven
growing,internationalinequalityinadultlongevitybetween1970and2000.
Whethertolookatdifferentwell-beingdimensionsofindividuallyortoresort
tomultidimensionalindicespresentsadilemma.Ontheonehand,theinterpretation
ofindividualindicesisstraightforwardandthatprovidesanadvantage.Ontheother,if
individualindicesshowconflictingtendencies,drawinggeneralconclusionsonits
evolutionbecomesimpossible(Decancqetal.,2009).Thishasledtoconstructing
6MorrisonandMurtin(2013)useBourguignonandMorrisson’strans-nationalunitsor“large”countries(32ratherthan33,intheircase).ItisworthnotingthatMorrisonandMurtinalsomeasurehumancapitalinequalitythatinsofaramonetarydimensionofeducationisbeyondthispaper’sscope.7GoeslingandFirebaugh(2004),onthebasisofsecondaryliterature,alsohypothesisedaninvertedUshapeevolutionofhealthinequalityoverthelasttwocenturies,startingfromlowlevelsthatwouldhaveincreasedsincethelatenineteenthcenturyandpeakedintheInterwaryears,todeclineduringthesecondhalfofthetwentiethcentury.
7
compositeindicatorsinwhichtwomainapproaches,welfareandcapabilities,canbe
distinguished.
Thepioneerinaddressinginternationalinequalitywithinacapabilities
framework,Ram(1992),noticedadiscrepancybetweenthehighlevelofincome
inequalityandthelowlevelofhumandevelopmentinequality.Earlier,onthebasisof
thePhysicalQualityofLifeIndex,Ram(1980)hadobservedasustaineddeclineinwell-
beinginequalityover1950-1970,atoddswiththesimultaneousrisingtrendobserved
forincome.Later,FarhadNoorbakhsh(2006)pointedoutaslowinequalityreduction
inhumandevelopmentduringthelastquarterofthetwentiethcentury.More
recently,RicardoMartínez(2012)hasfoundadeclineintheinternationaldispersionof
humandevelopmentbetween1980and2010.Intheonlylongrunperspective
availableonhumandevelopment,MorrissonandMurtin(2005)observedthatthe
evolutionofinequalityforaslightlymodifiedversionoftheUNDPHDIhadaninverted
U-shapewithaturningpointin1930.8
AlsoinspiredinAmartyaSen’scapabilitiesapproach,andonthebasisof
‘achievementindices’(seenextsection)fordifferentsocialindicators(infantmortality
rate,lifeexpectancyatbirth,anddailycalorieandproteinsupply),BartHobijnand
PhilipHansFranses(2001)disputedtheviewthatstandardsoflivingconvergedinthe
latetwentiethcenturysuggestinganincreaseinunweightedinequalitythatresulted
fromawideninggapbetweenCoreandPeripherysincethe1960s.9
Themainstylizedfactthatderivesfromthesurveyedliteratureis,therefore,a
long-termriseinwell--beinginequalitythatpeakedbytheearly-twentiethcentury
and,then,gavewaytoasustaineddecline.Thisisatoddswiththeevolutionof
internationalincomedistribution,inwhichdispersionroseuptoapeakby1950,
stabilisedand,then,declined.Canthisdepictionofthetrendsinwell-beinginequality
8Morerecently,Rijpma(2017)hasconstructedacomprehensivewell-beingindexforthelasttwocenturiesonthebasisofawiderangeofindicators(income,health,education,politicalinstitutions,freedom,inequality,andpersonalsecurity)andusingalatentvariablemodel.ThiscompositeindexshowsmoreintenseimprovementandstrongerconvergenceovertimethanGDPperhead.Also,Decancqetal.(2009)usingaflexibleindexofmultidimensionalwell-beingshowedadeclineinunweightedinequalityover1975-2000.9ThisconclusionwasdisputedbyNeumayer(2003),whorejectedHobijnandFranses‘achievementindices’toassessinequalityandusingtheoriginalvaluesofasetofsocialvariables(lifeexpectancy,infantsurvival,educationenrolment,literacy,andtelephoneandtelevisionavailability),restatedtheviewofareductionininequalitybetween1960and2000.
8
beconfirmedusingamorerigorousconceptualapproachandamorecomprehensive
database?
MeasuringWell-being
Howprogressinnon-economicdimensionsofwell-beingismeasured
constitutesafarfromnegligiblematter.Usuallyoriginalvaluesofsocialvariables(life
expectancy,height,orliteracy)areemployed(AcemogluandJohnson,2007;Beckeret
al.,2005;BourguignonandMorrisson,2002;MorrissonandMurtin,2013;Hattonand
Bray,2010;Lindert,2004).Non-incomewell-beingindicators,suchaslifeexpectancy,
height,infantmortality,literacyandenrolmentratesoryearsofschooling,have,unlike
GDP,asymptoticlimitsthatreflectbiologicalorphysicalmaxima.Thismeansthatthe
useoforiginalvaluesforcomparisonsoverspaceandtimeintroducesbiases,asthe
rangeofvariationisverynarrow,forcingsmallergains(bothabsoluteandrelative)as
theirlevelsgethigher(Sen,1981;Dasgupta,1990;CorniaandMenchini,2006).This
objectionisparticularlyrelevantwhenanattemptismadeatmeasuringthe
distributionofsuchvariableacrosscountriesandovertime,astheuseoforiginal
valuesunavoidablyleadstowardsconvergence.
Atransformationwouldbe,then,requiredtomeasurechangeswithinupper
andlowerbounds.Apossibilityisalineartransformation,suchastheoneusedinthe
UnitedNationsDevelopmentProgramme’s(UNDP)indexofhumandevelopment,
whichreducesthesizeofthedenominatorand,thus,widenstheindex’srange.Indices
foreachdimension(Ix)arecomputedaccordingas,
Ix=(x-Mo)/(M-Mo),[1]
xbeingtheobservedvalueofagivendimensionofwelfare,andMoandMthe
minimumandmaximumvalues(goalposts).Theindexvarieswithin0and1.
However,usinglinearlytransformedvaluesdoesnotsolvetheproblemas
absolutechangesofidenticalsizeresultinsmallermeasuredimprovementforthe
countrywiththehigherinitiallevel(Sen,1981;Kakwani,1993).10Consider,for
example,twoimprovementsinlifeexpectancyatbirth,onefrom30to40yearsand
10AsDecancqetal.(2009)putit,theproblemofspuriousconvergenceremains,nonetheless,withaconcavetransformationas“itdampenstheeffectofincreasingvaluesatthehigherendofthedistribution”(p.17).
9
anotherfrom70to80years.Theseincreasesareidenticalinabsoluteterms,butthe
secondissmallerinproportiontotheinitialstartinglevel.
Valuingequallyidenticalincreasesinabsolutetermsatlowandhighlevelsmay
raiseethicalobjections,though.However,asParthaDasgupta(1990:23)pointedout,
‘Equalincrementsarepossiblyoflessandlessethicalworthaslifeexpectancyrisesto
65or70yearsandmore.Butwearemeaningperformancehere.Soitwouldseem
thatitbecomesmoreandmorecommendableif,withincreasinglifeexpectancy,the
indexweretoriseatthemargin.Theideahereisthatitbecomesmoreandmore
difficulttoincreaselifeexpectancyaslifeexpectancyrises’.SuchviewissharedbySen
(1981:292),whoasserted,“aslongevitybecomeshigh,itbecomesmoreofan
achievementtoraiseitfurther”.Moreover,givingmoreweighttosavingthelifeof
youngeroverolderpeopleimpliesanarbitraryvaluejudgement(Deaton,2006).
Theshortcomingsofalineartransformationoforiginalvaluesbecomemore
evidentwhenqualityistakenintoaccount.Lifeexpectancyatbirth,andliteracyand
schoolingratesarejustcrudeproxiesfora“longandhealthylife”(Engineeretal.,
2009)andforaccesstoknowledge,respectively,whichconstitutethewell-beingaims
inthecapabilitiesapproach.
Medicalresearchconcludesthathealthylifeexpectancyincreasesinline(or
evenmorethanproportionally)withlifeexpectancyatbirthandthat,aslife
expectancyraises,disabilityforthesameage-cohortfalls(Salomonetal.,2012;
Mathersetal.,2001).Inotherwords,thequalityofliferisesforeachagecohortaslife
expectancyatbirthincreases.11Similarly,thequalityofeducation,measuredinterms
ofcognitiveskills,growsasthequantityofeducationincreases(HanushekandKimko,
11Thedeclineinage-specificdisabilityaslifeexpectancyatbirthincreasesiscompatible,however,withtherecentfindingthatyearslosttodisability(YLD)risewithlifeexpectancybecauseYLDtendtoconcentrateattheendoflife(Salomonetal.,2012).So,perhaps,theviewthatwhilelongevityincreases,periodsofill-healthcanbelonger,butarelivedinbetterhealthandlessdisability,duetomedicaltechnologicadvance(Manton,1982),qualifiesFries(1980)morbiditycompressionhypothesis(SeethediscussioninFriesetal.,2011andLindgren,2016).StudyingtheUnitedStatesover1990-2005Cutleretal.(2014)arguedthatthereductionindisabledlifeexpectancyandtheincreaseindisability-freelifeexpectancysuggestthecompressionofmorbidity.AlsoChernewetal.(2016)foundanincreaseinhealthylifeexpectancyalongafallindisablelifeexpectancyintheU.S.during1992-2008.However,thisfindingisnotconfirmedonthebasisofself-reportedchronicdiseaseandBeltrán-Sánchezetal.(2016)concludethatthereisnoclearevidenceofcompressionofmorbidity.ThescepticalviewalsofindsupportinthecaseofEurope,forwhichHegerandKolodziej(2016)donotfindthatmedicalprogressreducesthedisablingimpactofdiseases,associatingpopulationageingwithanextensionofmorbidity.
10
2000;Altinoketal.,2014).12Thebottomlineisthatmoreyearsoflifeandeducation
implyhigherqualityofhealthandeducationduringchildhoodandadolescence.
Unfortunatelynohistoricaldataonhealth-adjustedlifeexpectancyandquality-
adjustededucationmeasuresforthebroadcountrysampleconsideredhereare
availablebefore1990.Infact,whetheranassociationbetweenmortalityand
morbidityexistedbetween1870and1990remainsanunknown(Riley,1990;Howse,
2006;Bleakley,2007,2010).However,NanakKakwani’s(1993)proposalof
transformingsocialdimensionswithanonlinearfunction,inwhichachievementsof
thesameabsolutesizehavealargerimpactasthestartingpointishigher,mayprovide
ashort-cutmethodtoderiveproxiesforhealthylifeexpectancyandcognitiveskillson
thebasisofcrudefiguresforlifeexpectationatbirthandliteracyandenrolmentrates
andyearsofschooling.13
Kakwani(1993)constructedanormalisedindexfromanachievementfunction
inwhichanincreaseinthestandardoflivingofacountryatahigherlevelimpliesa
greaterachievementthanwouldhavebeenthecasehaditoccurredatalowerlevel,
f(x,Mo,M)=((M-Mo)1-ε–(M–x)1-ε)/((M-Mo)1-ε),for0<ε<1[2]
=f(x,Mo,M)=(log(M-Mo)–log(M–x))/log(M-Mo),forε=1[3]
wherexisanindicatorofacountry’sstandardofliving,MandMoarethemaximum
andminimumvalues,respectively,andlogstandsforthenaturallogarithm.
TheachievementfunctionproposedbyKakwani(1993)isaconvexfunctionof
x,anditisequalto0,ifx=Mo,andequalto1,ifx=M,ranging,thus,between0and
12Thecorrelationbetweenqualityandquantityofeducationover1965-2010appearshighatworldscalebutdeclineswhenthesampleisrestrictedtodevelopedcountries,suggestingthatasthequantityofeducationgetshigher,qualityincreasesbecomemorethanproportional(Altinoketal.,2014).13Lifeexpectancyatbirthdefined,bytheUN,as“Theaveragenumberofyearsthatanewborncouldexpecttolive,ifheorsheweresubjecttotheage-specificmortalityratesofagivenperiod”http://www.un.org/esa/sustdev/natlinfo/indicators/methodology_sheets/health/life_expectancy.pdf.Therateofadultliteracyisdefinedasthepercentageofthepopulationaged15yearsandoverwhoisabletoreadandwrite.Theunadjustedenrolmentrateisthepercentageofpopulationaged5-24enrolledinprimary,secondary,andtertiaryeducationthathasbeencorrectedforthepre-1980eratoobtaingrossenrolmentrates(GER).Since1980GERareavailable.Yearsofschoolingrepresenttheaverageyearsoftotalschooling(primary,secondary,andtertiary)forpopulationaged25andover.Seethediscussion,sources,andproceduresinPradosdelaEscosura(2018).
11
1.Inthiscontext,thelineartransformationrepresentsaparticularcase,forε=0,
whichyieldsexpression[1]foreachdimensionoftheindex.14
Well-beinginequalityisalsoaddresshereusingacompositeindex,which
representsanalternativetotheUNDP'sconventionalHDI,aHistoricalIndexofHuman
Development(HIHD)(PradosdelaEscosura,2015).AstheconventionalHDI,theHIHD
includesasproxiesforahealthylife,accesstoknowledge,andotheraspectsofwell-
being(notdirectlyassociatedtohealthandeducation),measuresoflongevity(life
expectancyatbirth),education(yearsofschooling),andmaterialwell-being
(discountedpercapitaincome,inlogs),respectively.Thedifferenceisthatnon-income
variablesaretransformednon-linearlyusingexpression[3],ratherthanlinearly(asin
theHDI),soincreasesofthesameabsolutesizerepresentgreaterachievementsthe
higherthelevelatwhichtheytakeplace.Asregardstheincomedimension,ithasbeen
derivedusingexpression[1]andpercapitaGDPinlogs.Althoughthisisafarfrom
satisfactorysolution,relaxingtheassumptionofdiminishingreturnstoincomewould
makepercapitaGDP–nothavinganasymptoticupperbound-thedriverofthehuman
developmentindex,renderingthelatterredundant.15
TheHIHDhasbeenobtainedasamultiplicativecombinationofthetransformed
valuesofeachdimension.16Ifwedenotethenon-linearlytransformedvaluesoflife
expectancyatbirthandyearsofschoolingasLandS,respectively,andtheadjusted
percapitaincomeasY,thehistoricalindexofhumandevelopmentisderivedas,
HIHD=L1/3S1/3Y1/3[4]
14Inthecaseoflifeexpectancy,maximumandminimumvaluesacceptedare85and20years,respectively,while0and100willbetheupperandlowerboundsforadultliteracyandgrossenrolment(primary,secondaryandtertiary)rates,and0and15yearsarethegoalpostsinthecaseofyearsofschooling(UNDP,2014).A‘floor’of25yearshasbeenassumedforoflifeexpectancyatbirth,whileforliteracyandenrolmentrates,thehighestandlowesthistoricalvalueshavebeensetat99and1percent,respectively,and0.01yearsinthecaseofyearsofschooling.Seethediscussion,sources,andcomputationproceduresinPradosdelaEscosura(2018).15SeeZambrano’s(2011a,2011b)theoreticaljustificationfortheintroductionofdiminishingreturnstoincome.Foradiscussionandanalternativeproposalexcludingthelogtransformationofincome,seeRavallion(2012)andHerreroetal.(2012).16Sincealldimensionsareconsideredindispensable,theyareassignedequalweights(UNDP,2010).Upperandlowerboundsandmaximumandminimumlevelsforlifeexpectancyandyearsofschoolingarethoseoffootnote14.Inthecaseofpercapitaincome,theupperandlowerbounds,expressedinGeary-Khamis[G-K]1990dollars,$46,949and$100,respectively.A‘floor’of$300hasbeenaccepted.SeethediscussioninPradosdelaEscosura(2018).
12
TrendsinWell-beingInequality
Iswell-beinginequalityhigherintheworldtodaythanitwasinthelate
nineteenthcentury?Canwedistinguishdifferentphasesinitsevolution?Howdo
differentdimensionsofwell-beingcompare?Whatwerethemaindriversof
inequality?
Theoverviewoftheliteratureshowsthatmoststudiesaddressinequalityin
socialdimensionsofwell-beingusingtheiroriginalvalues(and,occasionally,their
lineartransformation)butthisapproachtendstobiastheresultsfavouring
convergence.Suchspurioustendencyismitigatedhere,ifnottotallysuppressed,by
resortingtoKakwaninon-lineartransformationofhealthandeducationdimensions
(seeHobijnandFranses,2001).Therefore,theresultspresentedinthissectionshould
provideamoreaccuratepictureofwell-beinginequalitytrends.
Alternativeinequalitymeasuresarechosen,astheydifferintheirsensitivityto
differentpartsofthedistribution.Thus,AtkinsonclassA(ε)indices(inwhichε
representsaninequalityaversionparameterusuallyrangingfrom0.5to2.5)–withthe
largerε,themoresensitivetheindexistodifferencesatthebottomofthe
distribution-,forwhichIhavechosenε=2,isoneoption.TheGinicoefficient,whichis
moresensitivetothemiddle(mode)ofthedistribution,isalsoselected.Lastly,
entropyindicesG(0),whichcorrespondstoHenriTheil’s(1967)populationweighted
index,alsoknownasMeanLogarithmicDeviation(MLD),andG(1),orTheil’sincome
weightedindex,knownforshort,asTheil,areconsidered.InentropyindicesG(α),the
morepositive(negative)α,themoresensitivetheindexistodifferencesatthetop
(bottom)ofthedistribution.Hence,MLDismoresensitivetothebottom,andTheilto
thetopofthedistribution.
Twotypesofinequalityestimatesarepresentedhere,inequalitybetween
countryaveragesinwhichallcountriesaregiventhesameweight,regardlesstheir
size,thatis,Milanovic’s(2005)Inequality1;andinequalitybetweencountryaverages
butweightedbycountriessize,soalargecountrycountsmorethanasmallone,
namely,Milanovic’sInequality2.Theunweightedmeasureofinequality(Inequality1)
allowsforthefactthatpoliciesareimplementedatcountrylevelandimpactonits
citizens’well-being.Besides,weightedmeasures(Inequality2)areverysensitivetothe
13
performanceofhighlypopulatedcountries.However,Inequality2,althoughimplicitly
assumesperfectlyequalwithin-countrydistribution,doesgetusclosertoameasureof
worlddistributionbyassigninghighervaluetomorepopulatedcountries(Milanovic,
2005:7-8).Unfortunately,nodataonwithin-countrydistributionofsocialdimensions
ofwell-beingareavailableforsuchalargesampleandtimespan.17
Differentcountrysampleshavebeenusedinthealternativeinequality
estimates,forwhichthelongerthetimespan,thenarrowerthespatialcoverage.Thus,
overtheentiretimespan,1870-2015,96countriesareconsidered,itsnumberrising
upto105,138,155,and164countriesforthesamplesstartingin1913,1950,1980,
and1990,respectively.Thecountrysamplesrepresentabove90percentoftheworld
population.Theresultsofthesesampleshavebeensplicedusingasbenchmarkthe
morerecentperiod,whichhaslargercountrycoverage(PradosdelaEscosura,2018).
Doespopulation-weightedinternationalinequality(Inequality2)provideagood
proxyforglobalinequality,thatis,inequalityamongworldinhabitants(Inequality3)?
Thiswouldbethecaseifbetween-countryinequality,ratherthanwithin-country
inequality,drivesglobalinequality.18Evidenceonglobalinequalityestimatesbothfor
percapitaincomeandeducationsupportsthishypothesis(Bourguignonand
Morrisson,2002:vanZandenetal.,2013;MorrissonandMurtin,2013).Itwillbe,then,
assumedherethatinter-countrydispersionprovidesalowerboundmeasureofglobal
inequality.19
Asinternationalinequalityhasbeenusuallyassessedintermsofpercapita
GDP,Iwillstartbylookingatinequalityinaverageincomessotheconventional
yardstickisprovided.20Ifwefirstlydelveintounweightedmeasures(Inequality1),a
sustainedincreaseinincomeinequalityappearsforthoseindicessensitiveto
17ClioInfrahttps://www.clio-infra.eu/effortstoprovideinequalitymeasuresforeducationandincomefallshortoftheamountofdatarequiredinmyestimates.18Thisassertiononlyappliesinthecontextofperfectlydecomposableentropyindices,butfortheGini,theoverlapcomponent,whichtendstoevolveinoppositedirectiontobetween-countryinequality,alsomatters(Milanovic,2005:25).19Itcouldalsobearguedthatthedispersioninsocialdimensionstendstobesignificantlowerthaninthecaseofincome,namely,thelongevityoreducationgapbetweentherichandthepoorislessthanproportionaltotheirincomegap.20ThedatasourcesandproceduresareexposedinPradosdelaEscosura(2018).Thecountryestimatesforthedifferentwell-beingdimensionsexaminedhereareaccessibleathttp://espacioinvestiga.org/home-hihd/?lang=en.
14
changesatthebottomofthedistribution:intheAtkinsoncoefficient,A(ε=2),allthe
wayto2000,stabilisinghereafter;alsointheMLD,butforflatteningbetweenthe
1950sand1980s,andfallingwiththenewcentury(Figure1aandTable1).
Meanwhile,fortheTheil,ameasuresensitivetothetopofthedistribution,therise
ininequalitystoppedin1950,openingthewaytoasteadydeclinedownto1990
andshadowingMLDbehaviourhenceforth.TheGini,moresensitivetothemiddle
ofthedistribution,alsopeakedin1950,stabilisingthereafterbutfluctuatingalong
theTheil.Itisworthstressingthesharpincreaseininequalitybetweentheendof
theGreatDepressionand1950acrossallinequalitymeasuresthatmaybe
attributedtotheunevenimpactofWorldWarII.
Whenpopulation-weightedinequalityacrosscountries(inequality2),is
examined,allindicesshowasustainedriseupto1950,slowingdowntothe1980s,
anddecliningthereafter(Figure1bandTable1).Indices’behaviour,nonetheless,
differedsincethemid-twentiethcentury,withtheGiniandtheTheil,ratherstable
andonlyfallingsince2000,andtheMLDandAtkinsonrisingandfalling,with1980
asaturningpoint.
PopulationandGDPexpandedatdifferentpaceacrosscountriesoverthelong
run.DidinequalityrisebecausepercapitaincomeGDPgrewatdifferentratesacross
countries,orjustbecausepopulationgrewfasterincountrieswitheitherloworhigh
income?Awaytoprovideanansweristosimulatetheyearlyratesatwhich,other
thingsbeingequal,inequalitywouldhaveevolvedhadallcountriesenjoyedidentical
percapitaincome(population)growth.21Theresultsfromthesimulationindicatethat
differencesinthepaceofeconomicgrowthacrosscountriesexplaintheevolutionof
incomeinequality,whilenosignificantimpactderivesfromthedispersioninratesof
populationgrowth(Table2).
Thereisaconsensusintheliteratureabouttheimpactoflargecountrieson
population-weightedinequality,buthowmuchdotheyconditionInequality2levels
andtrends?ExcludingIndiaandChinaallowustofindouthowmuchofthe
describedtrendsininternationalincomeinequalityresultsfromtheirevolution 21Thepracticalwayofcarryingoutthesimulationherehasbeencomputingweightedinequalitymeasuresinwhichinitiallevelofpopulationhasbeenkeptconstantoverspecificperiods:1870-1913,1913-1950,1950-1990,and1990-2015.
15
(Figure2andTable3).ItappearsthatthetwolargeAsiancountriescontributed
significantlytoahigherlevelofinternationalinequalityuntil1990,particularlyover
1913-1980,andtoitsreductionthereafter.Alternatively,excludingSub-Saharan
Africaallowsustoobservethat,sincethemid-1980s,thisregionhasgradually
contributedtoraisinginternationalincomeinequality.
Afurtherquestionistheextenttowhichinequality2isdrivenbythegap
betweenadvancedanddevelopingcountriesandbythedispersionwithineachof
thesetwogroups.HereadvancedcountriescomprisethosebelongingtotheOECD
priortoitsenlargementin1994(OECD,forshort,ortheWest,asthisgroupof
countriesisindistinctivelylabelledthroughoutthepaper).22Itcanbenoticedthatthe
gapbetweentheWestandtheRestwasthemaincontributortotheaggregatelevelof
inequalityuntilthemid-1950s,onthebasisoftheMLD,andfortheentiretimespan
consideredintermsoftheTheil.Thedispersionwithinthetwogroupsexplainstrends
ininternationalinequalityover1930-2000,boththesustainedrisebetweentheearly
1930sandmid-1970sanditsreductionfrom1980to2000(Figures3aand3band
Table4).Acloserlookshowsthatitwasthegrowingdisparitybetweendeveloping
regionswhatdeterminedthelevelsandtrendsofpost-1950internationalincome
inequality(Table4,col.6).
Doesinternationalinequalityinsocialdimensionsofwell-beingreplicatetrends
andlevelsofincomeinequality?Accesstoknowledgeisacomponentofwelfaretobe
considered.Threealternativemeasuresofeducationareconsideredhere:theliteracy
rate,astockvariablethatiscertainlythemostrelevantoneinearlystagesof
developmentbutthathasalsobeenconsideredasthemaindriverofthedispersion
overthelongrunofamorecompleteindicator,yearsofschooling(Morrisonand
Murtin,2013).Thegrossenrolmentrateforprimary,secondary,andtertiary
education,aflowvariablethatprobablycapturesbetterformaleducationand,since
22Inthispaper,OECD,ortheWest,coincideslargely–butnotcompletely–withthemembershipoftheOrganisationforEconomicCo-operationandDevelopment(OECD)upto1994:WesternEurope,its'WesternOffshoots',andJapan.WesternEuropeincludesAustria,Belgium,Denmark,Finland,France,Germany,Greece,Ireland,Italy,theNetherlands,Norway,Portugal,Spain,Sweden,SwitzerlandandtheUnitedKingdom.ThreeOECDmembersareexcluded:IcelandandLuxemburgareleftoutforlackofhumandevelopmentestimates;TurkeyiscountedwithAsiainordertoreducegroupheterogeneityintermsofdevelopment.'WesternOffshoots'consistsofAustralia,Canada,NewZealandandtheUnitedStates.
16
mostofitsprogresshasbeenachievedthroughpubliceducation,providesameasure
ofgovernmentintervention.Lastly,yearsofschooling,anotherstockvariable,isthe
mostcomprehensivemetricbuthas,nonetheless,thedownsideofitslowercountry
coverage.23
HighlevelsofinequalityIandII-wellabovethoseforpercapitaincomeandlife
expectancyatbirth-areobservedforallaccesstoknowledgeproxymeasuresinthe
latenineteenthandearlytwentiethcentury,priortothediffusionofmassprimary
education(BenavotandRiddle,1988;Lindert,2004).
Inthecaseofliteracy,bothInequalityIandIIstartedfromhighlevelsand
experiencedsteadydeclinethroughoutmostofthetwentiethcentury,eventhough
thelevelsremainedhighuntilitslastquarter(Figures4aand5b).Forexample,theGini
movedfrom0.7to0.3over1870-2015butwasstillat0.5in1970andonlyfellbelow
0.4by1990.Moreover,whenmeasuredbytheAtkinsonindex,inequalityremained
stableuntilmid-twentiethcenturyfollowing,then,alongrunandsteadydecline(Table
5).
Intermsofenrolment,inequalityfelltotheearly1980s,butsufferedareversal
inthe1990s,stabilisingthereafter(Figures5aand5b).OnlyinthecaseoftheAtkinson
index,thedeclineofInequality1wasdeferreduntilthesecondhalfofthetwentieth
century(Table6).Asinthecaseofliteracy,inequalitylevelsremainedhighuntilmid-
twentiethcentury(itonlyfellbelow0.4inthe1950s).
InequalityinyearsofschoolingfellsteadilysinceWorldWarI,exceptforthe
Atkinsonindexthatinitiateditsdeclineafterwards(since1938and1929forthe
unweightedandpopulation-weightedmeasures,respectively)(Figures6aand6b).
Nonetheless,inequalitywasstillhighinthethirdquarterofthetwentiethcenturywith
aGiniof0.45aslateas1975(Table7).
Inalleducationmeasures,thedispersionofitsimprovementacrosscountries,
ratherthanthevarianceofpopulationgrowth,explainstheinequalitycontractionover
thelongrun(Table8).
Inordertoascertaintheextenttowhichlargecountriesconditioned
population-weightedmeasuresofinequality,Ihavesimulatedliteracy,enrolment,and
23AlsothelessdetaileddiscussionofthedatasetsourcesbytheirauthorsLeeandLee(2016)andBarroandLee(2013)hasbeentakenintoaccount.SeePradosdelaEscosura(2018).
17
yearsofschoolinginequalityintheabsenceofChinaandIndia,firstly,and,then,inthe
absenceoftheSub-SaharanAfricanregion(Figures7a,7b,and7c).Itappearsthat
ChinaandIndiacontributedtohigherlevelsofinequalityintheInterwaryearswhen
measuredintermsofliteracy;priortoWorldWarII,intermsofenrolment;and
between1913and1960,inyearsofschooling;and,then,tolowerlevelsinthe1960s
andfrom1990onwards(literacy),andsince1950and1970(enrolmentandyearsof
schooling,respectively).Interestingly,ChinaandIndiaalsocontributedtolower
inequalityintermsofliteracyandyearsofschoolingpriorto1900.Sub-SaharanAfrica
contributedtohigherinequalitysince1890(literacy)and1950(enrolment),andfrom
the1920sonwards(yearsofschooling),butalsotolowerinequalitypriorto1900in
thecaseofenrolment(Tables9-11).
HowdotrendsineducationinequalitycomparetothoseobtainedbyMorrisson
andMurtin(2013)?Theseauthorscomputedinter-countryeducationinequalityonthe
basisoftheoriginalvaluesofaverageyearsofschoolingand,therefore,theresulting
levelofinequalityissystematicallylowerthanthoseprovidedhere,asbounded
variablesexhibitaspurioustendencytowardsconvergence.Nonetheless,inequality
fellinMorrissonandMurtin’sestimates,asalsohappenedwhenIreplicatedthe
measurementofinequalityusingoriginalvaluesoftheeducationvariables(literacy
andgrossenrolmentratesandyearsofschooling).Nonetheless,thereductionof
inequalitywasmuchlessintensethanwhencomputedonthebasisonKakwaniindices
and,inthecaseofenrolment,noreversalappearsinthe1990s(Figures8a,8b,and
8c).
Breakingdownaggregateinequality2intoinequalitybetweenOECDcountries
andtherestoftheworldandinequalitywithineachofthetwogroupsrevealsthat,in
thecaseofliteracy,thegapbetweentheWestandtheRestmadethemain
contributiontothelevelofinternationalinequalityuntiltheearly1930s,accordingto
theTheil,andhadasimilarcontributiontothatofthedispersionwithineachgroup
until1900accordingtotheMLD;butlargelydrovetotalinequalityuntilmid-twentieth
century(WorldWarII,MLD;1950,Theil).Henceforth,thedispersioninliteracy
standardswithinthetwogroupswasthemaindriver(Table12andFigures9aand9b).
Acloserinspectionrevealsthatitwasactuallythedispersionwithinthedeveloping
regionswhatdroveinequalitytrendssincemid-twentiethcentury(Table12,col.6).
18
Inthecaseofenrolment,thegapbetweentheWestandtheRestrepresented
thelargestshareofinternationalinequalityuntilthetwentiethcentury(1900forthe
MLDand1930fortheTheil)andonlydroveinequalitytrendsuntilthe1920s,when
within-groupinequality-largelywithintheRest-tookover,exceptforthe1990s(Table
13andFigures10aand10b).Thus,althougheducationalpracticesbecameincreasingly
globalisedsincetheinterwaryears,werediffusedunevenlyindevelopingregions
(Table13,col.6).
Whenmeasuredinyearsofschooling,boththelevelandtrendofinequalityare
determinedbythedispersionwithintheWestandtheRest(Table14),andmore
specificallythelatter,accordingtotheMLD,butonlysince1950,onthebasisofthe
Theil(Figures11aand11b).
Longevityisanessentialwell-beingdimension.Inequalitytrendsinlife
expectancy,asmeasuredbydifferentunweightedindices,provideacommonpicture,
althoughlessintensewhencomputedbytheGinicoefficient(Table15).Different
phasescanbeidentifiedforunweightedandpopulation-weightedinequality.
Inequality1roseuptotheearly1920s,butforareversalbetween1900and1913.
Then,asustainedinequalitydeclinetookplacebetweenthelate1920sandmid-1980s,
moreintenseduringthelatethirtiesandforties.Thedecliningtrendrevertedafter
1990,withinequalitygoingbacktoitslevelofthemid-1970sbythemid-2000s,
stabilisinglaterandreturningtothelevelofthelate1990sby2015(Figure12a).Inthe
caseofInequality2,theevolutionwassimilarbutformoreintenseincreaseuptothe
mid-1920swhenhigherlevelswerereached,andasharpercontractionto1985,
followedbyamilderreversalupto2005(Figure12b).
Theperformanceoflargecountrieshadafarfromnegligibleimpacton
inequality2(Table16).IndiaandChinacontributedsignificantlytoriseofinequalityup
tothemid-1920sand,then,partlyoffsetitsreductionuntilthemid-1950s.In
particular,thehighmortalityinChinaduringMao’sGreatLeapForwardcausedthe
dramaticincreaseininequalitybroughtinequalitybacktothepre-WorldWarIIhigh
levelby1960.Sincethe1960s,ChinaandIndiahavecontributedtotheinequality
decline(Figure13).ExcludingSub-SaharanAfricadoesnotalterworldinequality
trends,butreducesitsleveluptotheearly1920sandincreasesitfromthelate1920s
onwards,especially,sincethemid-1960s.
19
Itisalsoworthnotingthatthevarianceinlifeexpectancygainsacross
countries,ratherthanthedispersioninpopulationgrowth,iswhatexplainsthe
inequalitydeclineoverthelongrun(Table17).
Howdoinequalitytrendsinlifeexpectancycomparetothoseobtainedby
previousstudies?BourguignonandMorrisson(2002)computedinequality2onthe
basisoftheoriginalvaluesoflifeexpectancyatbirthand,sinceboundedvariables
tendtoconvergeovertime,thelevelofinequalityissystematicallylowerthaninmy
estimates;nonetheless,theirestimatesalsoexhibitasustaineddecline.Thisinequality
underestimateisconfirmedbythecomparisonbetweenlifeexpectancyinequality
measurescomputedfromoriginalvaluesandKakwaniindices.Furthermore,the
inequalitymeasureobtainedfromoriginalvaluesdoesnotcapturethepost-1985
reversal(Figure14).
Trendsininequality2havebeendecomposedintobetween-groupinequality,
thatis,thegapbetweentheOECDandtheRestoftheworld,andinequalitywithin
eachofthesetwogroups(Table20).ItisonlyuptoWorldWarII,thatthegap
betweentheWestandtheRestmadethelargestcontributiontointernational
inequality;thereafter,withingroupinequalitybecameitsmaincontributor(Figures
15aand15b).However,thedispersionwithinOECDandtheRestdroveinequality
trendsuptothe1920s,whilethereductioninthegapbetweentheWestandtheRest
steeredthedeclineinaggregateinequalitybetween1929and1970.Thepost-1990
increaseininequalitywasduetothedispersionwithintheRest(Table18).
Sofartrendsinlifeexpectancyinequalityhavebeendrawn,buthowcanthey
beexplained?Andwhyaresodifferentfromthoseofpercapitaincome?
Healthimprovementscanbedepictedintermsofahealthfunction(Preston,
1975).Movementsalongthefunctionrepresentgainsthatcanbeattributedto
economicgrowthandresultinnutritionimprovements-thatstrengthentheimmune
systemandreducemorbidity(Stolnitz,1955;McKeownetal.,1962,1975;Fogel,
2004)-andinthepublicprovisionofhealth(Loudon,2000;CutlerandMiller,2005).
Outwardshiftsinthehealthfunctioncapture,inturn,medicaltechnologicalchange,
whichhasbeenthemaincontributortothelongrunincreaseinlongevitysincethe
latenineteenthcentury(McKinlayandMcKinlay,1977;Riley,2005b;Cutleretal.,
2006).
20
Theepidemiologicalorhealthtransition–thatis,thephaseinwhichpersistent
gainsinlowermortalityandhighersurvivalareachievedasinfectiousdiseasegives
waytochronicdiseaseasthemaincauseofdeath(Omran,1971;Riley,2001)-is
associatedtothediffusionofthegermtheoryofdiseasesincethe1880s(Preston,
1975)thatledtotheintroductionofnewvaccinessincethe1890s.Moreover,medical
technologicalprogressintroducednewdrugstocureinfectiousdiseases-sulphadrugs
sincethelate1930s(Easterlin1999)andantibioticssincethe1950s-,contributingto
spreadthehealthtransition(Easterlin,1999;Jayachandranetal.,2010;Lindgren,
2016).
However,widespreadgainsinlongevitythatlaybeneaththedeclineinlife
expectancyinequalityduringtheglobalisationbacklash(1914-1950),atimeof
economicdistress,demandfurtherexplanationsthatarealsolinkedtothegerm
theoryofdisease.Thus,improvementsinpublichealth–oftenatlowcost,aslow
incomespreventedthepurchaseofthenewdrugs-andthediffusionofpreventive
methodsofdiseasetransmissionandknowledgedisseminationthroughschool
educationcontributedtoreducinginfantmortalityandmaternaldeath,twomajor
determinantsoftheincreaseinlifeexpectancyatbirthindevelopingregions(Riley,
2001).
Duringtheepidemiologicaltransitionsubstantialachievementsinlongevity
wereattainedbutnotsharedequallywithinsocietiesandacrosscountries.Lackof
economicmeansandbasicscientificknowledgepreventedafastandwidediffusionof
newmedicaltechnologyandhealthpracticeacrosscountries.Thus,inthelate
nineteenthandearlytwentiethcentury,theincreaseinlifeexpectancyinequalitycan
beassociatedtothefactthatthefirsthealthtransitionwasunevenlydistributedand
initiallyrestrictedtoadvancedwesterncountries.
Thegradualinternationaldiffusionofthehealthtransitionbetweenthe1920s
andthe1970shelpstoexplainthereductioninlifeexpectancyinequality.Its
contractionwasparticularlyintenseduringthe1930sand1940s,atatimeofstagnant
ordecliningaverageincomesinmanycountriesandincreasinginternationalincome
inequalityasaresultoftheGreatDepressionandWorldWarII(Table1andFigures2a
and2b).Thiswasduetotheimprovementoflifeexpectancyincountriesoflowper
21
capitaincomelevelsasaconsequenceofthefirsthealthtransition(Mandle,1970;
Reher,2003;Riley,2005a).
Theincreaseinlifeexpectancyinequalityafter1990maybeassociated,notjust
totheimpactofHIV-AIDSinSubSaharanAfricaortothedemiseofsocialisminEastern
Europe,butalsotoasecondhealthtransitionthatsofarhasbeenrestrictedtothe
West.Inthisnewtransition,mortalityandmorbidityfallamongtheelderlyasaresult
ofnewmedicalknowledgethathaspermittedabettertreatmentofrespiratoryand
cardiovasculardiseaseandvisionproblems(particularly,cataractsurgery)(Cutleret
al.,2006;Chernewetal.,2016;EgglestonandFuchs,2012).Theriseinlongevityhas
alsobeenhelpedbybetternutritioninearlyyearsoflife.Theresultispeoplelivingnot
justlongerlifebutlongerhealthylifeyears(Mathersetal.,2001;MurrayandLopez,
1997;Salomonetal.,2012).24
Tosumup,episodesofrisinginternationalinequalityinlifeexpectancyinthe
latenineteenthandearlytwentiethcenturyand,then,inthelatetwentiethandearly
twentieth-firstcentury,coincidewiththeearlystagesoftheepidemiologicalandthe
secondhealthtransitions,respectively,andresultfromtheunevendiffusionofnew
medicalknowledgeandtechnologyandhealthpractices.
Thus,anassociationcouldbeproposedbetweentheleveloflongevityandits
internationaldistribution.TheGinithatcorrespondstodifferentlevelsoflife
expectancyatbirth(bothestimatedfromKakwaniindices)ispresentedinFigure16a.
AninvertedUshapecurverelationshipresultsthatcouldbedeemedaHealthKuznets
Curve.25ThedrivingforceoftheHealthKuznetsCurve(HKC)wouldbethespreadof
thehealthtransitionsacrosscountries.Theunevendiffusionoftheepidemiologicalor
firsttransitionwouldaccountforitsrise,whileitsgradualdiffusionacrosstheglobe
wouldexplainitsdecline.ItisworthnotingthattheHKCshowsasteeprisereachinga
peakatlowlevelsoflifeexpectancyandasmootherandlongerdecline.
24Politicsmayhavealsocontributedtoshapeinequalitytrends.Ithasbeenarguedthatautocraciessuccessfullyfoughtmortalitythroughgovernmentinterventionduringthefirsthealthtransition,butdemocraciesfacilitatedthenewhealthtransitionwithfewerconstraintsonmedicalinnovation,sogainsinlifeexpectancyhavebeenlargerindemocraticthaninauthoritariancountriessincethelatetwentiethcentury(Devereux,2010;Mackenbach,2013).25ThisdepictionoftheHealthKuznetsCurvediffersfromtheoneproposedbyCosta-Fontetal.(2018)whocompareincome-relatedhealthinequalityandpercapitaincome.Inmycase,bothlevelsandinequalitymeasurescorrespondtolifeexpectancy.
22
Moreover,itcouldbeinsinuatedthattheclosingoftheHealthKuznetsCurve
gavewaytoanewone.ThenewHKCappears,nonetheless,tohavebeenshort-lived
aslongevitydifferencesacrosscountriesresultingfromtheunevendiffusionofthe
secondhealthtransition(Cutleretal.2006)havebeenoffset,atleasttemporarily,by
therecoveryoflifeexpectancyinSubSaharanAfricaandinformersocialistEurope.
Thesameexercisehasbeenreplicatedonthebasisofuntransformedvaluesof
lifeexpectancyatbirthwithsimilarresultsbutfortheemergenceofasecondHKC
(Figure16b).
Sincethedifferentwell-beingdimensionsdiscussedexhibitcontradictory
trends,alookattheinternationaldistributionofacompositemeasuresuchasthe
HistoricalIndexofHumanDevelopment(HIHD)seemswarranted.26
Theevolutionofhumandevelopmentinequality1showsthat,afteraphaseof
stabilityinthelatenineteenthcentury,butfortheincreaseoftheAtkinsonindex,it
experiencedasustaineddeclinebetweenthe1900sand1980sand,then,stabilisedfor
therestofthetwentiethcentury,resumingitsdeclinesince2010(Figure17aand
Table19).InthecaseofInequality2,itroseupto1900forthoseindicesmore
sensitivetochangesatthebottominthedistribution(MLDandAtkinson)andthe
subsequentdeclinewasmoreintenseacrosstheboarduptothe1980sand,then,
continuedatsteady,butslower,paceuntil2015(Figure17b).
Itisworthnotingthat,asinthecaseofitsdimensions,itisthevariancein
humandevelopmentgainsacrosscountries,ratherthanthedispersioninpopulation
growth,whatexplainedinequalityevolution(Table20).
LargecountriesaffectedsignificantlytheevolutionofInequality2.Chinaand
Indiacontributedtoraisingitsleveluntil1960andreducingitsincethemid-1970s
(Figure18).SubSaharanAfricahasmadeasustainedcontributiontoraisinginequality
sincethemid-1960s,especiallyafter1990(Table21).
Whenaggregateinequality2isdecomposedintothegapbetweentheWest
andtheRestandthedispersionwithinbothgroups,itappearsthatthegapmadethe
largestcontributiontothelevelofinternationalinequalityupto1900fortheMLDand
26TheHIHDusedhereisanadaptationofthe2010(UNDP,2010)indexandusesaverageyearsoftotalschoolingforpopulationaged25andoverasthemeasureofeducation.SeethediscussionofthealternativeHIHDestimatesinPradosdelaEscosura(2018).
23
the1930sfortheTheil(Figures190aand19bandTable22),whilethedispersion
withinthetwogroups,especiallyintheRest,tookoverfrommid-twentiethcentury
onwards(Table26).However,inequalitytrendsweredrivenbythegapbetweenthe
WestandtheRestuptotheearly1960s.
Giventhemultiplicativecompositionofthehumandevelopmentindex
(expression4),whenmeasuredwithanentropyindex,abreakdownofhuman
developmentinequalitycanbeperformedintotheequallyweightedsumofeach
components’inequality-lifeexpectancy(L),yearsofschooling(S),andadjusted
income(Y)-,plusaresidual,thataccountsforthedisparitiesbetweenthecomponents’
distributions(Martínez,2016:417-418).
MLDhihd=1/3MLDL+1/3MLDS+1/3MLDY+R[5]
Theilhihd=1/3TheilL+1/3TheilS+1/3TheilY+R[6]
Itcanbeobservedthatthelevelofhumandevelopmentinequalitydepended
chieflyonthedistributionofeducationoverthelongrun,whilelongevityraised
inequalityuptothemid-1920sanditscontributionfadedawaysincetheearly1960s
(Figures20aand20bandTable23).
ConcludingRemarks
Well-beinginequalityhasdeclinedovertime.Theglobalisationofmassprimary
educationandthehealthtransitionsappearasthemaindriversofsuchanequalising
trend.ThegapbetweentheWestandtheRestexplainsonlypartiallytheevolutionof
well-beinginequalityasthedispersionwithindevelopingregionsincreasingly
determineditstrends.
Thediffusionofthehealthtransitionshasdrivenlifeexpectancyinequality.
Duringmostofthetwentiethcentury,astheepidemiologicaltransitionspreadacross
theglobe,lifeexpectancyroseanditsdispersionfell.Then,itcametoahaltasthe
transitionwascompleted.Attheturnofthecentury,anew,secondhealthtransition,
linkedtothesuccessfulfightagainstcardiovascularandrespiratorydisease,emerged,
extendinglifeexpectancyoftheelderlyinthedevelopedworldandprovokingan
increaseinhealthinequality.
24
Thesefindingsareatoddswiththeviewonlongruninequalityderivedfrom
realpercapitaGDP.Whilepopulation-weightedincomeinequalityincreaseduntilthe
thirdquarterofthetwentiethcentury,inequalityinsocialdimensionsdeclinedsince
WorldWarI.Furthermore,thecontrastbetweeninequalityintermsofincomeand
humandevelopmentisstriking(Figure21)andchallengestheideathatpercapita
incomeprovidesagoodpredictorofwelfaretrends.
Whyinequalitydeclinedintermsofsocialdimensions,butnotofGDPper
head?Wasitduetopublicpolicy,ortothefactthatmedicaltechnologyisapublic
good?WhyhastherebeennosecondhealthtransitionintheRest?Isittheoutcome
ofinequalisingnewmedicaltechnologies,oroflackofpublicpolicies?Asthenew
medicaltechnologiesbecomeaccessible,isitforeseeableadeclineinlifeexpectancy
inequalityinthetwentieth-firstcentury?Answeringthesechallengingquestions
requiresanambitiousresearchprogramme.
25
ReferencesAcemoglu,D.andS.Johnson(2007),“DiseaseandDevelopment:TheEffectsofLife
ExpectancyonEconomicGrowth,”JournalofPoliticalEconomy115:925-985.Altinok,N.,C.Diebolt,andJ.L-Demeulemeester(2014),“ANewInternationalDataseton
EducationQuality,1965-2010”,AppliedEconomics46(11):1212-1247.Anand,S.andA.Sen(2000),“TheIncomeComponentoftheHumanDevelopmentIndex,”
JournalofHumanDevelopment1:83-106. Barro,R.J.andJ.W.Lee(2013),“ANewDataSetofEducationalAttainmentintheWorld,
1950–2010”,JournalofDevelopmentEconomics104:184-198.Becker,G.S.,T.J.Philipson,andR.R.Soares(2005),TheQuantityandQualityofLifeandthe
evolutionofworldinequality,AmericanEconomicReview95(1):277–291Beltrán-SánchezH.,M.P.Jiménez,andS.V.Subramanian(2016),“AssessingMorbidity
CompressioninTwoCohortsfromtheHealthandRetirementStudy”,JournalofEpidemiologyandCommunityHealth70(10):1011–1016
Benavot,A.andP.Riddle(1988),“TheExpansionofPrimaryEducation,1870-1940:TrendsandIssues,”SociologyofEducation61:191-210.
Berry,A.,F.BourguignonandC.Morrisson(1983),“ChangesintheWorldDistributionbetween1950and1977”,EconomicJournal93(37):331-350
Bleakley,H.(2007),“DiseaseandDevelopment:EvidencefromHookwormEradicationintheAmericanSouth,”QuarterlyJournalofEconomics122:73-117.
Bleakley,H.(2010),“MalariaEradicationintheAmericas:ARetrospectiveAnalysisofChildhoodExposure”,AmericanEconomicJournal:AppliedEconomics2:1-45.
Bourguignon,F.andC.Morrisson(2002),“InequalityamongWorldCitizens”,AmericanEconomicReview92,4:727-744.
Chernew,M.,D.M.Cutler,K.Gosh,andM.B.Landrum(2016),UnderstandingtheImprovementinDisabilityFreeLifeExpectancyintheU.S.ElderlyPopulation,NBERWorkingPaperSeries22306.
ClioInfrahttps://www.clio-infra.eu/Cornia,G.A.andL.Menchini(2006),HealthimprovementsandHealthInequalityduringthe
Last40Years,UNU-WIDERResearchPaper2006/10.Costa-Font,J.,C.Hernández-Quevedo,andA.Sato(2018),“AHealth‘Kuznets’Curve’?Cross-
SectionalandLongitudinalEvidenceonConcentrationIndices”,SocialIndicatorsResearch136:439-452.
Cutler,D.M.andG.Miller(2005),“TheRoleofPublicHealthImprovementsinHealthAdvance:TheTwentiethCenturyUnitedStates”,Demography42:1-22.
Cutler,D.M.,A.DeatonandA.Lleras-Muney(2006),“TheDeterminantsofMortality”,JournalofEconomicPerspectives20:97-120.
Cutler,D.M.,K.Ghosh,andM.B.Landrum(2014),EvidenceforSignificantCompressionofMorbidityIntheElderlyU.S.Population,inD.A.Wise(ed.),DiscoveriesintheEconomicsofAging,UniversityofChicagoPress,pp.21-51.
Dasgupta,P.(1990),“Well-BeingandtheExtentofitsRealizationinPoorCountries”,EconomicJournal100(400):1-32.
Deaton,A.(2006),“GlobalPatternsofIncomeandHealth:Facts,Interpretations,andPolicies,”NBERWorkingPapers12735.
Deaton,A.(2013),TheGreatEscape.Health,Wealth,andtheOriginsofInequality,Princeton:PrincetonUniversityPress.
26
Decancq,K.,A.Decoster,andE.Schokkaert(2009),“TheEvolutionofWorldInequalityinWell-being”,WorldDevelopment37(1):11-25.
Devereux,J.(2010),TheHealthoftheRevolution:ExplainingtheCubanHealthcareParadox,QueensCollege,CUNY.
Dowrick,S.andM.Akmal(2005),“ContradictoryTrendsinGlobalIncomeInequality:ATaleofTwoBiases”,ReviewofIncomeandWealth51(2):201-230.
Easterlin,R.A.(1974),“Doeseconomicgrowthimprovehumanlot?”,inP.A.DavidandM.W.Reder(eds.),NationsandHouseholdsinEconomicGrowth:EssaysinHonorofMosesAbramovitz,NewYork:AcademicPress,pp.89-125.
Easterlin,R.(1999),“HowBeneficientistheMarket?ALookattheModernHistoryofMortality”,EuropeanReviewofEconomicHistory3(3):257-294.
Edwards,R.D.(2011),“ChangesinWorldInequalityinLengthofLife:1970-2000”,PopulationandDevelopmentReview37(3):499-528.
Eggleston,K.N.andV.Fuchs(2012),“TheNewDemographicTransition:MostGainsinLifeExpectancyNowRealizedLateinLife”,JournalofEconomicPerspectives26:137-156.
Engineer,M.H.,N.RoyandS.Fink(2010),“’Healthy’HumanDevelopmentIndices,”SocialIndicatorsResearch99:61-80.
Firebaugh,G.(1999),“EmpiricsofWorldIncomeInequality”,AmericanJournalofSociology104(6):1597-1630.
Fogel,R.W.(2004),TheEscapefromHungerandPrematureDeath,1700-2010.Europe,AmericanandtheThirdWorld,CambridgeUniversityPress,NewYork.
Fries,J.F.(1980),“Aging,naturaldeath,andthecompressionofmorbidity”,NewEnglandJournalofMedicine303(3):130–135.
Fries,J.F., Fries,B.Bruce,andE.Chakravarty(2011),“CompressionofMorbidity1980–2011:AFocusedReviewofParadigmsandProgress”,JournalofAgingResearchhttps://www.hindawi.com/journals/jar/2011/261702/
GallardoAlbarrán,D.(2017),Missedopportunities?TheDevelopmentofHumanWelfareinWesternEurope,1913-1950,GroningenGrowthandDevelopmentCentreResearchMemorandum166.
Goesling,B.andD.P.Baker(2008),“Threefacesofinternationalinequality”,ResearchinSocialStratificationandMobility26:183–198
Goesling,B.andG.Firebaugh(2004),“TheTrendinInternationalHealthInequality”,PopulationandDevelopmentReview30(1):131-146.
Hanushek,E.A.andD.D.Kimko(2000),“Schooling,Labor-ForceQuality,andtheGrowthofNations”,AmericanEconomicReview90:1184-1208.
Hatton,T.J.andB.E.Brey(2010),“LongRunTrendsintheHeightsofEuropeanMen,19th–20thCenturies”,EconomicsandHumanBiology8:405-413.
Heger,D.andI.W.K.Kolodziej(2016),ChangesinMorbidityoverTime:EvidencefromEurope,RuhrEconomicPapers,No.640.
Herrero,C.,R.Martínez,andA.Villar(2012),“ANewerHumanDevelopmentIndex”,JournalofHumanDevelopmentandCapabilities13:247–268
Hobijn,B.andP.H.Franses(2001),“AreLivingStandardsConverging?”,StructuralChangeandEconomicDynamics12:171-200.
Howse,K.,(2006)“IncreasingLifeExpectancyandtheCompressionofMorbidity:aCriticalReviewoftheDebate,”OxfordInstituteofAgeingWorkingPaper226.
Ivanov,A.andM.Peleah(2010),“FromCentrallyPlannedDevelopmenttoHumanDevelopment,”UNDPHumanDevelopmentReportsResearchPaper38
27
Jayachandran,S.,A.Lleras-Muney,andK.V.Smith(2010),“ModernMedicineandtheTwentiethCenturyDeclineinMortality:EvidenceontheImpactofSulfaDrugs,”AmericanEconomicJournal:AppliedEconomics2:118-146.
Jones,C.I.andP.J.Klenow(2016),“BeyondGDP?WelfareacrossCountriesandTime”,AmericanEconomicReview106(9):2426-2457
Kahneman,D.,andA.Deaton(2010),HighIncomeImprovesEvaluationofLifebutnotEmotionalWell-being,ProceedingsoftheNationalAcademyofSciencesoftheUnitedStatesofAmerica107(38):16489–16493.
Kakwani,N.(1993),“PerformanceinLivingStandards.AnInternationalComparison”,JournalofDevelopmentEconomics41:307-336.
Korzeniewicz,R.P.andT.P.Moran(1997),“WorldEconomicTrendsintheDistributionofIncome,1965-1992”,AmericanJournalofSociology102(4):1000-1039
Lee,J.W.andH.Lee(2016),“HumanCapitalintheLongRun”,JournalofDevelopmentEconomics122:147–169
Liberati,P.(2015),“TheWorldDistributionofIncomeAnditsInequality,1970–2009”,ReviewofIncomeandWealth61(2):248-273.
Lindert,P.H.(2004),GrowingPublic.SocialSpendingandEconomicGrowthsincetheEighteenthCentury,Cambridge:CambridgeUniversityPress.
Lindgren,B.(2016),TheRiseinLifeExpectancy,HealthTrendsamongtheElderly,andtheDemandforCare.ASelectedLiteratureReview,NBERWorkingPaper22521
Loudon,I.(2000),“MaternalMortalityinthePastanditsRelevancetoDevelopingCountriesToday”,AmericanJournalofClinicalNutrition72,(1)(supplement):241S-246S.
Mackenbach,J.P.(2013),“PoliticalConditionsandLifeExpectancyinEurope,1900-2008”,SocialScience&Medicine82:134-146.
Maddison,A.(2006),TheWorldEconomy,OECDDevelopmentCentre,Paris,2006.MaddisonProject(2013), TheMaddison-Project,http://www.ggdc.net/maddison/maddison-
project/home.htmMandle,J.R.(1970),“TheDeclineofMortalityinBritishGuiana,1911-1960,”Demography7:
301-315.Manton,K.G.(1982),“ChangingConceptsofMorbidityandMortalityintheElderly
Population”,MilbankMemorialFundQuarterly/HealthandSociety60:183-244.Martínez,R.(2012),“InequalityandtheNewHumanDevelopmentIndex”,AppliedEconomics
Letters19(6):533-535.Martínez,R.(2016),“InequalityDecompositionandHumanDevelopment”,JournalofHuman
DevelopmentandCapabilities17(3):415-425.Mathers,C.D.,R.Sadana,J.A.Salomon,C.J.L.Murray,andA.D.Lopez(2001),“HealthyLife
Expectancyin191Countries”,Lancet357:1685–1691.McKeown,T.andR.G.Record(1962),“ReasonsfortheDeclineofMortalityinEnglandand
WalesduringtheNineteenthCentury”,PopulationStudies16:94-122.McKeown,T.,R.G.Record,andR.D.Turner(1975),“AnInterpretationoftheDeclineof
MortalityinEnglandandWalesduringtheTwentiethCentury”,PopulationStudies29:391-422.
McKinlay,J.B.andS.M.McKinlay(1977),“TheQuestionableContributionofMedicalMeasuretotheDeclineofMortalityintheUnitedStatesintheTwentiethCentury”,MilbankMemorialFundQuarterlyHealthandSociety55:405-428
Milanovic,B.(2005),WorldsApart.MeasuringInternationalandGlobalInequality.Princeton:PrincetonUniversityPress.
28
Milanovic, B. (2016), Global Inequality: A New Approach for the Age of Globalization,Cambridge,MA:BelknapPressofHarvardUniversityPress.
Milanovic,B.andJ.E.Roemer(2016),“InteractionofGlobalandNationalIncomeInequalities”,JournalofGlobalizationandDevelopment7(1):109–115.
Morrisson,C.andF.Murtin(2005),“TheWorldDistributionofHumanCapital,LifeExpectancyandIncome:AMulti-dimensionalApproach”(mimeo).
Morrison,C.andF.Murtin(2009),“TheCenturyofEducation”,JournalofHumanCapital3(1):1-42
Morrisson,C.andF.Murtin (2013), “TheKuznetsCurveofHumanCapital Inequality:1870–2010”,JournalofIncomeInequality11,3:283-301.
Murray,C.J.L.andA.D.Lopez(1997),“RegionalPatternsofDisability-freeLifeExpectancyandDisability-adjustedLifeExpectancy:GlobalBurdenofDiseaseStudy”,Lancet349:1347-1352.
Neumayer,E.(2003),“BeyondIncome:ConvergenceinLivingStandards,BigTime”,StructuralChangeandEconomicDynamics14:275-96
Nordhaus,W.,andJ.Tobin(1972),“IsGrowthObsolete?”,inNationalBureauofEconomicResearch,EconomicGrowth:FiftiethAnniversaryColloquiumV,NewYork:ColumbiaUniversityPress,pp.1–80
Noorbakhsh,F.(2006),InternationalConvergenceorHigherInequalityinHumanDevelopment?Evidencefor1975to2002,UNU-WIDERResearchPaperNo.2006/15
Omran,A.R.(1971),“TheEpidemiologicalTransition:ATheoryofEpidemiologyofPopulationChange,”MilbankMemorialFundQuarterly49(4):509-538.
Pinker,S.(2018),EnlightenmentNow.TheCaseforReason,Science,HumanismandProgress,LondonandNewYork:AllenLane.
PradosdelaEscosura,L.(2015),“WorldHumanDevelopment:1870-2007”,ReviewofIncomeandWealth61,2:220-247.
PradosdelaEscosura,L.(2018),HumanDevelopmentinHistoricalPerspective:NewEstimates,UniversidadCarlosIIIWorkingPapersinEconomicHistory18-06.Datasetaccessibleathttps://espacioinvestiga.org/home-hihd/?lang=en
Preston,S.(1975),“TheChangingRelationshipbetweenMortalityandtheLevelofEconomicDevelopment”,PopulationStudies29(2):231-248.
Ram,R.(1980),“PhysicalQualityofLifeIndexandInter-countryInequality”,EconomicLetters5:195-199.
Ram,R.(1992),“InternationalInequalitiesinHumanDevelopmentandRealIncome”,EconomicLetters38:351-354.
Ram,R.(2006),“StateoftheLifeSpanRevolutionbetween1980and2000”,JournalofDevelopmentEconomics80:518-526.
Ravallion,M.(2012),“TroublingTradeoffsintheHumanDevelopmentIndex”,JournalofDevelopmentEconomics99:201–209
Reher,D.S.(2003),“PerfilesdemográficosdeEspaña1940-1960”,inC.Barciela(ed.),AutarquíayMercadonegro.Elfracasoeconómicodelprimerfranquismo,1939-1959,Barcelona:Crítica,pp.1-26
Riley,J.C.(1990),“TheRiskofBeingSick:MorbidityTrendsinFourCountries”,PopulationandDevelopmentReview16:403-432.
Riley,J.C.(2001),RisingLifeExpectancy.AGlobalHistory,NewYork:CambridgeUniversityPress.
29
Riley,J.C.(2005a),PovertyandLifeExpectancy.TheJamaicaParadox,CambridgeUniversityPress,NewYork
Riley,J.C.(2005b),“TheTimingandPaceofHealthTransitionsAroundtheWorld”,PopulationandDevelopmentReview31(4):741-764
Rijpma,A.(2017),WhatCan’tMoneyBuy?WellbeingandGDPsince1820,CGEHWorkingPaperSeries78
Sala-i-Martin,X.(2006),“TheWorldDistributionofIncome:FallingPoverty...andConvergence,Period”,QuarterlyJournalofEconomics121(2):351-397.
Salomon,J.A.,H.Wang,M.K.Freeman,T.Vos,A.D.Flaxman,A.D.Lopez,andC.J.L.Murray(2012),“HealthyLifeExpectancyfor187Countries,1990–2010:ASystematicAnalysisfortheGlobalBurdenDiseaseStudy2010,”Lancet380:2144-2162.
Schultz,T.P.(1998),“Inequalityinthedistributionofpersonalincomeintheworld:Howitischangingandwhy”,JournalofPopulationEconomics11:307-344.
Sen,A.K.(1981),“PublicActionandtheQualityofLifeinDevelopingCountries,”OxfordBulletinofEconomicsandStatistics43:287-319.
Stolnitz,G.J.(1955),‘ACenturyofInternationalMortalityTrends’,PopulationStudies9(1):24-55.Theil,H.(1967),EconomicsandInformationTheory,Amsterdam:NorthHolland.UnitedNationsDevelopmentProgramme[UNDP](2014), HumanDevelopmentReport2014.
SustainingHumanProgress:ReducingVulnerabilitiesandBuildingResilience,NewYork:UnitedNationsDevelopmentProgramme.
Veenhoven,R.,&Vengust,F.(2013),“TheEasterlinIllusion:EconomicGrowthDoesGowithGreaterHappiness,InternationalJournalofHappinessandDevelopment1(4),311–343
Zambrano,E.(2011a),AnAxiomatizationoftheHumanDevelopmentIndex,UNDPHumanDevelopmentReportsResearchPaper10
Zambrano,E.(2011b),Functionings,Capabilitiesandthe2010HumanDevelopmentIndex,UNDPHumanDevelopmentReportsResearchPaper11
Zanden,J.L.van,J.Baten,P.Foldvári,andB.vanLeeuwen(2014),“TheChangingShapeofGlobalInequality1820-2000;ExploringaNewDataset”,ReviewofIncomeandWealth60(2):279-297.
30
Table1
InternationalInequalityinRealPerCapitaGDP,1870-2015
Unweighted
Populationweighted
MLD Theil Gini Atkinson(ε=2) MLD Theil Gini Atkinson(ε=2)1870 0.26 0.32 0.40 0.36 0.17 0.20 0.32 0.251880 0.29 0.36 0.42 0.39 0.21 0.25 0.36 0.301890 0.31 0.38 0.44 0.42 0.25 0.28 0.39 0.341900 0.33 0.40 0.45 0.44 0.29 0.32 0.43 0.391913 0.37 0.44 0.47 0.48 0.35 0.36 0.46 0.451925 0.39 0.45 0.48 0.50 0.38 0.40 0.48 0.481929 0.42 0.47 0.49 0.53 0.40 0.41 0.49 0.501933 0.38 0.44 0.47 0.49 0.32 0.33 0.44 0.441938 0.42 0.48 0.49 0.53 0.40 0.39 0.48 0.511950 0.48 0.54 0.52 0.58 0.56 0.54 0.55 0.621955 0.48 0.53 0.52 0.58 0.53 0.51 0.54 0.601960 0.47 0.49 0.51 0.58 0.55 0.51 0.54 0.621965 0.46 0.44 0.50 0.59 0.57 0.52 0.55 0.621970 0.48 0.45 0.51 0.61 0.60 0.54 0.56 0.641975 0.48 0.41 0.50 0.62 0.62 0.54 0.56 0.661980 0.48 0.40 0.49 0.63 0.62 0.55 0.56 0.661985 0.48 0.39 0.49 0.63 0.59 0.56 0.56 0.641990 0.51 0.42 0.50 0.65 0.62 0.59 0.57 0.641995 0.56 0.46 0.52 0.68 0.57 0.57 0.56 0.632000 0.58 0.47 0.53 0.70 0.56 0.57 0.55 0.622005 0.58 0.45 0.52 0.71 0.50 0.50 0.52 0.612010 0.54 0.42 0.50 0.70 0.41 0.40 0.47 0.582015 0.53 0.41 0.50 0.70 0.38 0.36 0.45 0.58
31
Table2
PopulationGrowthDispersion:ContributiontoRealPerCapitaGDPInequality
(population-weightedMLDandTheil)
MLD
Theil
Actual FixedInitialPopulation Actual FixedInitialPopulation
1870 0.18 0.18 0.21 0.211880 0.22 0.22 0.25 0.251890 0.26 0.25 0.28 0.281900 0.31 0.30 0.33 0.321913 0.36 0.35 0.37 0.37
1913 0.36 0.36 0.37 0.371925 0.40 0.39 0.41 0.401929 0.42 0.42 0.42 0.411933 0.34 0.34 0.34 0.341938 0.42 0.43 0.40 0.401950 0.57 0.58 0.55 0.53
1950 0.57 0.57 0.55 0.551955 0.55 0.55 0.53 0.521960 0.56 0.57 0.52 0.521965 0.58 0.59 0.54 0.531970 0.61 0.62 0.56 0.541975 0.63 0.64 0.56 0.531980 0.63 0.65 0.57 0.541985 0.61 0.63 0.57 0.541990 0.62 0.64 0.59 0.55
1990 0.62 0.62 0.59 0.591995 0.57 0.57 0.57 0.562000 0.56 0.55 0.57 0.552005 0.50 0.48 0.50 0.482007 0.47 0.45 0.46 0.432010 0.41 0.39 0.40 0.372015 0.38 0.35 0.36 0.32
32
Table3
InternationalInequalityinRealGDPperhead,1870-2015:TheContributionofChinaandIndiaandSubSaharanAfrica
(population-weightedMLD)
Excluding Excluding
Actual ChinaandIndia SubSaharanAfrica
1870 0.17 0.17 0.171880 0.21 0.21 0.211890 0.25 0.22 0.241900 0.29 0.23 0.291913 0.35 0.25 0.351925 0.38 0.29 0.381929 0.40 0.29 0.401933 0.32 0.23 0.321938 0.40 0.26 0.401950 0.56 0.39 0.561955 0.53 0.39 0.531960 0.55 0.39 0.551965 0.57 0.42 0.561970 0.60 0.44 0.601975 0.62 0.45 0.611980 0.62 0.46 0.611985 0.59 0.50 0.571990 0.62 0.56 0.581995 0.57 0.61 0.522000 0.56 0.63 0.502005 0.50 0.61 0.422010 0.41 0.56 0.322015 0.38 0.55 0.28
33
Table4
DecomposingInternationalInequalityinRealPerCapitaGDP,1870-2015(population-weightedMLDandTheil)
MLD
Within-group Between-group TOTAL OECD TheRest1870 0.03 0.14 0.17 0.05 0.031880 0.04 0.17 0.21 0.05 0.041890 0.06 0.19 0.25 0.05 0.061900 0.08 0.21 0.29 0.05 0.091913 0.11 0.23 0.35 0.05 0.131925 0.12 0.26 0.38 0.06 0.131929 0.14 0.26 0.40 0.05 0.161933 0.12 0.20 0.32 0.04 0.141938 0.18 0.23 0.40 0.06 0.201950 0.27 0.29 0.56 0.12 0.301955 0.25 0.28 0.53 0.09 0.281960 0.28 0.27 0.55 0.06 0.321965 0.29 0.28 0.57 0.05 0.341970 0.34 0.26 0.60 0.03 0.401975 0.37 0.24 0.62 0.03 0.431980 0.38 0.24 0.62 0.02 0.431985 0.34 0.25 0.59 0.03 0.391990 0.34 0.28 0.62 0.02 0.391995 0.30 0.27 0.57 0.02 0.342000 0.28 0.28 0.56 0.02 0.312005 0.26 0.24 0.50 0.03 0.292010 0.24 0.18 0.41 0.02 0.262015 0.23 0.15 0.38 0.03 0.26
Theil
Within-group Between-group TOTAL OECD TheRest
1870 0.04 0.16 0.20 0.04 0.031880 0.04 0.20 0.25 0.04 0.041890 0.06 0.22 0.28 0.05 0.071900 0.07 0.25 0.32 0.05 0.101913 0.09 0.27 0.36 0.05 0.131925 0.09 0.30 0.40 0.05 0.151929 0.10 0.30 0.41 0.05 0.171933 0.09 0.24 0.33 0.03 0.151938 0.13 0.27 0.39 0.05 0.211950 0.20 0.34 0.54 0.11 0.301955 0.17 0.34 0.51 0.08 0.291960 0.18 0.33 0.51 0.05 0.331965 0.19 0.33 0.52 0.04 0.351970 0.22 0.32 0.54 0.03 0.421975 0.24 0.30 0.54 0.02 0.441980 0.24 0.31 0.55 0.02 0.441985 0.23 0.32 0.56 0.03 0.411990 0.23 0.36 0.59 0.02 0.431995 0.21 0.36 0.57 0.02 0.382000 0.19 0.37 0.57 0.02 0.352005 0.18 0.32 0.50 0.02 0.302010 0.16 0.23 0.40 0.02 0.252015 0.16 0.20 0.36 0.03 0.23
34
Table5
InternationalInequalityinLiteracy,1870-2015(KakwaniIndices)
Unweighted
Populationweighted
MLD Theil Gini Atkinson(ε=2) MLD Theil Gini Atkinson(ε=2)1870 1.45 1.06 0.73 0.94 1.14 0.87 0.67 0.921880 1.46 1.02 0.72 0.95 1.03 0.84 0.67 0.891890 1.48 1.00 0.72 0.96 1.02 0.84 0.67 0.891900 1.44 0.94 0.70 0.96 1.04 0.81 0.67 0.911913 1.42 0.92 0.69 0.96 1.02 0.79 0.66 0.891925 1.30 0.81 0.66 0.96 1.00 0.76 0.65 0.871929 1.26 0.79 0.65 0.95 0.99 0.75 0.65 0.871933 1.20 0.76 0.64 0.95 0.95 0.71 0.63 0.861938 1.17 0.73 0.63 0.95 0.85 0.64 0.60 0.841950 1.02 0.64 0.60 0.92 0.69 0.50 0.54 0.811955 0.86 0.58 0.57 0.87 0.61 0.46 0.51 0.761960 0.78 0.54 0.55 0.84 0.56 0.43 0.50 0.731965 0.67 0.48 0.52 0.79 0.50 0.39 0.48 0.691970 0.56 0.42 0.50 0.73 0.46 0.38 0.47 0.631975 0.51 0.38 0.47 0.69 0.43 0.36 0.46 0.591980 0.43 0.33 0.45 0.63 0.38 0.34 0.45 0.541985 0.38 0.30 0.42 0.59 0.34 0.30 0.42 0.501990 0.33 0.25 0.39 0.54 0.28 0.25 0.39 0.431995 0.29 0.23 0.37 0.50 0.25 0.23 0.37 0.412000 0.25 0.19 0.34 0.45 0.22 0.19 0.34 0.382005 0.23 0.18 0.32 0.43 0.20 0.18 0.33 0.352010 0.20 0.16 0.30 0.41 0.17 0.15 0.30 0.322015 0.19 0.14 0.29 0.38 0.16 0.14 0.29 0.31
35
Table6
InternationalInequalityinGrossEnrolment,1870-2015(KakwaniIndices)
Unweighted
Populationweighted
MLD TheilGiniAtkinson(ε=2) MLD TheilGiniAtkinson(ε=2)1870 0.98 0.77 0.64 0.79 1.42 1.12 0.76 0.931880 0.99 0.75 0.64 0.81 1.43 1.05 0.74 0.951890 0.95 0.68 0.61 0.81 0.94 0.79 0.66 0.831900 0.95 0.65 0.60 0.82 1.05 0.80 0.67 0.871913 0.87 0.59 0.58 0.81 0.92 0.69 0.63 0.851925 0.71 0.48 0.52 0.79 0.62 0.52 0.55 0.721929 0.66 0.44 0.51 0.77 0.55 0.46 0.52 0.681933 0.63 0.43 0.50 0.76 0.56 0.46 0.51 0.691938 0.60 0.40 0.49 0.77 0.62 0.50 0.54 0.741950 0.48 0.32 0.43 0.73 0.33 0.27 0.41 0.611955 0.41 0.28 0.40 0.69 0.25 0.20 0.35 0.541960 0.32 0.24 0.38 0.57 0.19 0.15 0.30 0.431965 0.25 0.20 0.34 0.47 0.17 0.14 0.30 0.361970 0.22 0.17 0.32 0.44 0.15 0.13 0.28 0.301975 0.20 0.15 0.30 0.41 0.13 0.11 0.26 0.281980 0.16 0.12 0.27 0.35 0.11 0.09 0.24 0.221985 0.14 0.11 0.25 0.32 0.11 0.10 0.25 0.211990 0.15 0.12 0.26 0.31 0.12 0.12 0.27 0.221995 0.17 0.15 0.29 0.33 0.14 0.15 0.28 0.242000 0.19 0.18 0.32 0.35 0.16 0.16 0.30 0.262005 0.18 0.18 0.33 0.31 0.14 0.15 0.30 0.232010 0.17 0.17 0.32 0.28 0.13 0.14 0.28 0.212015 0.17 0.17 0.32 0.27 0.12 0.13 0.27 0.20
36
Table7
InternationalInequalityinYearsofSchooling,1870-2015(KakwaniIndices)
Unweighted
Populationweighted
MLD TheilGiniAtkinson(ε=2) MLD TheilGiniAtkinson(ε=2)1870 0.90 0.58 0.57 0.95 0.81 0.57 0.58 0.901880 0.95 0.60 0.58 0.97 1.32 0.89 0.70 1.001890 0.96 0.61 0.58 0.98 1.38 0.87 0.69 1.021900 0.95 0.59 0.58 0.98 1.09 0.78 0.66 0.951913 1.04 0.63 0.59 1.00 1.39 0.92 0.71 1.031925 0.90 0.56 0.57 0.97 1.16 0.78 0.66 0.991929 0.82 0.53 0.55 0.93 1.01 0.71 0.63 0.951933 0.79 0.51 0.55 0.93 0.88 0.65 0.62 0.901938 0.74 0.49 0.54 0.91 0.79 0.61 0.60 0.861950 0.65 0.46 0.52 0.84 0.62 0.52 0.55 0.741955 0.60 0.43 0.50 0.80 0.53 0.46 0.52 0.681960 0.55 0.40 0.49 0.77 0.48 0.43 0.51 0.641965 0.52 0.38 0.48 0.73 0.42 0.38 0.48 0.591970 0.49 0.37 0.47 0.71 0.39 0.36 0.47 0.571975 0.44 0.34 0.45 0.67 0.36 0.34 0.45 0.531980 0.38 0.29 0.42 0.62 0.32 0.31 0.42 0.481985 0.32 0.24 0.39 0.56 0.26 0.26 0.39 0.421990 0.28 0.22 0.36 0.52 0.23 0.23 0.37 0.381995 0.25 0.20 0.35 0.48 0.21 0.21 0.35 0.352000 0.24 0.19 0.34 0.45 0.19 0.19 0.33 0.322005 0.23 0.18 0.33 0.43 0.18 0.18 0.32 0.302010 0.22 0.18 0.33 0.41 0.17 0.17 0.32 0.292015 0.20 0.16 0.32 0.37 0.15 0.15 0.29 0.25
37
Table8
PopulationGrowthDispersion:ContributionEducationInequality(population-weightedMLD)
Literacy
GrossEnrolment YearsofSchooling
Actual
FixedInitialPopulation Actual
FixedInitialPopulation Actual
FixedInitialPopulation
1870 1.03 1.03 1.25 1.25 0.78 0.781880 0.93 0.92 1.27 1.27 1.27 1.271890 0.92 0.90 0.83 0.82 1.33 1.321900 0.95 0.92 0.93 0.93 1.05 1.021913 0.92 0.90 0.81 0.82 1.33 1.34
1913 0.92 0.92 0.81 0.81 1.33 1.331925 0.90 0.90 0.54 0.53 1.11 1.091929 0.90 0.89 0.48 0.47 0.97 0.961933 0.86 0.85 0.49 0.49 0.84 0.841938 0.77 0.75 0.55 0.54 0.76 0.761950 0.67 0.59 0.33 0.26 0.61 0.58
1950 0.67 0.67 0.33 0.33 0.61 0.611955 0.59 0.59 0.24 0.24 0.52 0.521960 0.55 0.55 0.19 0.19 0.47 0.471965 0.49 0.49 0.17 0.17 0.41 0.411970 0.45 0.46 0.15 0.15 0.39 0.391975 0.42 0.43 0.13 0.13 0.36 0.361980 0.37 0.38 0.11 0.10 0.31 0.311985 0.33 0.34 0.11 0.11 0.26 0.261990 0.28 0.28 0.12 0.13 0.23 0.23
1990 0.28 0.28 0.12 0.12 0.23 0.231995 0.25 0.25 0.14 0.14 0.21 0.212000 0.22 0.21 0.16 0.15 0.19 0.182005 0.20 0.19 0.14 0.14 0.18 0.172007 0.19 0.18 0.13 0.13 0.17 0.162010 0.17 0.16 0.13 0.13 0.17 0.162015 0.16 0.15 0.12 0.12 0.15 0.14
38
Table9
InternationalInequalityinLiteracy,1870-2015:
TheContributionofChinaandIndiaandSubSaharanAfrica(MLD)(KakwaniIndices)
Excluding Excluding
Actual ChinaandIndia SubSaharanAfrica
1870 1.14 1.12 1.151880 1.03 1.10 1.021890 1.02 1.11 1.001900 1.04 1.08 1.011913 1.02 0.97 0.971925 1.00 0.85 0.961929 0.99 0.82 0.951933 0.95 0.78 0.901938 0.85 0.76 0.791950 0.69 0.69 0.581955 0.61 0.62 0.521960 0.56 0.59 0.481965 0.50 0.52 0.431970 0.46 0.47 0.411975 0.43 0.43 0.381980 0.38 0.39 0.351985 0.34 0.35 0.311990 0.28 0.30 0.251995 0.25 0.28 0.232000 0.22 0.25 0.202005 0.20 0.23 0.182010 0.17 0.20 0.142015 0.16 0.19 0.13
39
Table10
InternationalInequalityinGrossEnrolment,1870-2015:TheContributionofChinaandIndiaandSubSaharanAfrica(MLD)(KakwaniIndices)
Excluding Excluding
Actual ChinaandIndia SubSaharanAfrica
1870 1.42 1.05 1.511880 1.43 0.98 1.521890 0.94 0.90 0.971900 1.05 0.79 1.081913 0.92 0.68 0.931925 0.62 0.47 0.611929 0.55 0.44 0.531933 0.56 0.43 0.551938 0.62 0.47 0.601950 0.33 0.38 0.271955 0.25 0.30 0.191960 0.19 0.25 0.141965 0.17 0.23 0.131970 0.15 0.19 0.111975 0.13 0.16 0.091980 0.11 0.13 0.091985 0.11 0.13 0.101990 0.12 0.14 0.111995 0.14 0.17 0.122000 0.16 0.19 0.132005 0.14 0.17 0.132010 0.13 0.16 0.122015 0.12 0.16 0.11
40
Table11
InternationalInequalityinYearsofSchooling,1870-2015:TheContributionofChinaandIndiaandSubSaharanAfrica(MLD)(KakwaniIndices)
Excluding Excluding
ActualChinaandIndiaSubSaharanAfrica
1870 0.81 1.47 0.821880 1.32 1.44 1.341890 1.38 1.39 1.411900 1.09 1.26 1.071913 1.39 0.83 1.401925 1.16 0.69 1.151929 1.01 0.63 1.001933 0.88 0.59 0.861938 0.79 0.55 0.771950 0.62 0.50 0.601955 0.53 0.48 0.501960 0.48 0.46 0.451965 0.42 0.43 0.391970 0.39 0.41 0.371975 0.36 0.40 0.341980 0.32 0.35 0.291985 0.26 0.30 0.241990 0.23 0.29 0.221995 0.21 0.26 0.192000 0.19 0.25 0.172005 0.18 0.23 0.152010 0.17 0.22 0.152015 0.15 0.19 0.13
41
Table12
DecomposingInternationalInequalityinLiteracy(population-weighted):MLDandTheil(KakwaniIndices)
MLD
Within-group Between-group TOTAL OECD TheRest1870 0.61 0.48 1.14 0.19 0.691880 0.51 0.47 1.03 0.18 0.581890 0.49 0.48 1.02 0.17 0.551900 0.58 0.42 1.04 0.15 0.681913 0.54 0.43 1.02 0.13 0.641925 0.58 0.39 1.00 0.09 0.701929 0.62 0.35 0.99 0.08 0.751933 0.63 0.30 0.95 0.07 0.761938 0.63 0.22 0.85 0.07 0.761950 0.54 0.16 0.69 0.05 0.651955 0.49 0.13 0.61 0.04 0.581960 0.47 0.10 0.56 0.04 0.561965 0.40 0.11 0.50 0.03 0.471970 0.37 0.10 0.46 0.02 0.431975 0.34 0.09 0.43 0.02 0.391980 0.30 0.08 0.38 0.02 0.351985 0.27 0.07 0.34 0.01 0.311990 0.21 0.07 0.28 0.01 0.231995 0.19 0.06 0.25 0.01 0.222000 0.17 0.05 0.22 0.00 0.202005 0.16 0.05 0.20 0.00 0.182010 0.14 0.03 0.17 0.00 0.162015 0.14 0.02 0.16 0.00 0.16
Theil
Within-group Between-group TOTAL OECD TheRest
1870 0.27 0.54 0.87 0.14 0.491880 0.27 0.51 0.84 0.13 0.501890 0.26 0.52 0.84 0.13 0.521900 0.30 0.47 0.81 0.12 0.631913 0.27 0.47 0.79 0.10 0.591925 0.29 0.43 0.76 0.07 0.691929 0.33 0.39 0.75 0.06 0.751933 0.34 0.34 0.71 0.06 0.731938 0.38 0.25 0.64 0.05 0.731950 0.32 0.18 0.50 0.04 0.561955 0.31 0.15 0.46 0.03 0.511960 0.32 0.12 0.43 0.03 0.501965 0.27 0.13 0.39 0.02 0.421970 0.26 0.12 0.38 0.02 0.401975 0.25 0.11 0.36 0.02 0.371980 0.24 0.10 0.34 0.01 0.341985 0.22 0.09 0.30 0.01 0.301990 0.16 0.09 0.25 0.01 0.221995 0.15 0.07 0.23 0.01 0.202000 0.13 0.05 0.19 0.00 0.172005 0.12 0.05 0.18 0.00 0.162010 0.11 0.03 0.15 0.00 0.142015 0.11 0.03 0.14 0.00 0.14
42
Table13DecomposingInternationalInequalityinGrossEnrolment(population-weighted):
MLDandTheil(KakwaniIndices)
MLD
Within-group Between-group TOTAL OECD TheRest
1870 0.51 0.79 1.42 0.09 0.611880 0.69 0.66 1.43 0.08 0.841890 0.41 0.47 0.94 0.06 0.491900 0.57 0.43 1.05 0.06 0.701913 0.55 0.35 0.92 0.04 0.681925 0.39 0.22 0.62 0.06 0.471929 0.34 0.20 0.55 0.04 0.411933 0.40 0.16 0.56 0.04 0.481938 0.56 0.08 0.62 0.04 0.691950 0.24 0.09 0.33 0.04 0.281955 0.18 0.06 0.25 0.04 0.221960 0.16 0.04 0.19 0.04 0.181965 0.15 0.03 0.17 0.07 0.161970 0.12 0.03 0.15 0.02 0.141975 0.11 0.02 0.13 0.01 0.131980 0.09 0.02 0.11 0.00 0.111985 0.09 0.02 0.11 0.02 0.101990 0.09 0.03 0.12 0.03 0.101995 0.08 0.06 0.14 0.06 0.092000 0.10 0.06 0.16 0.03 0.112005 0.09 0.05 0.14 0.03 0.102010 0.07 0.06 0.13 0.03 0.082015 0.07 0.05 0.12 0.02 0.07
Theil
Within-group Between-group TOTAL OECD TheRest
1870 0.22 0.81 1.12 0.08 0.731880 0.28 0.69 1.05 0.07 0.931890 0.21 0.52 0.79 0.05 0.561900 0.27 0.48 0.80 0.05 0.731913 0.27 0.39 0.69 0.04 0.661925 0.25 0.25 0.52 0.06 0.471929 0.22 0.23 0.46 0.04 0.401933 0.26 0.19 0.46 0.03 0.471938 0.42 0.09 0.50 0.04 0.671950 0.16 0.11 0.27 0.04 0.241955 0.12 0.07 0.20 0.03 0.171960 0.11 0.05 0.15 0.04 0.141965 0.11 0.03 0.14 0.06 0.131970 0.10 0.03 0.13 0.02 0.131975 0.09 0.02 0.11 0.01 0.111980 0.08 0.02 0.09 0.00 0.101985 0.08 0.03 0.10 0.02 0.091990 0.08 0.04 0.12 0.03 0.101995 0.08 0.07 0.15 0.06 0.082000 0.09 0.07 0.16 0.03 0.112005 0.09 0.07 0.15 0.03 0.102010 0.07 0.07 0.14 0.03 0.092015 0.07 0.06 0.13 0.02 0.08
43
Table14DecomposingInternationalInequalityinYearsofSchooling(population-weighted):
MLDandTheil(KakwaniIndices)
MLD
Within-group Between-group TOTAL OECD TheRest
1870 0.63 0.20 0.81 0.29 0.941880 0.86 0.46 1.32 0.28 1.121890 0.90 0.48 1.38 0.25 1.211900 0.59 0.48 1.09 0.19 0.771913 0.86 0.53 1.39 0.13 1.051925 0.76 0.40 1.16 0.10 0.931929 0.67 0.34 1.01 0.09 0.821933 0.58 0.30 0.88 0.09 0.711938 0.53 0.27 0.79 0.09 0.631950 0.41 0.21 0.62 0.08 0.491955 0.34 0.18 0.53 0.08 0.401960 0.31 0.17 0.48 0.08 0.361965 0.27 0.15 0.42 0.09 0.301970 0.25 0.14 0.39 0.10 0.281975 0.23 0.13 0.36 0.11 0.261980 0.21 0.11 0.32 0.11 0.221985 0.18 0.09 0.26 0.10 0.191990 0.15 0.08 0.23 0.08 0.161995 0.14 0.07 0.21 0.07 0.152000 0.13 0.06 0.19 0.05 0.142005 0.11 0.06 0.18 0.04 0.122010 0.10 0.06 0.17 0.04 0.112015 0.09 0.06 0.15 0.03 0.10
Theil
Within-group Between-group TOTAL OECD TheRest
1870 0.41 0.17 0.57 0.26 0.931880 0.46 0.42 0.89 0.25 1.171890 0.42 0.44 0.87 0.22 1.111900 0.33 0.44 0.78 0.19 0.801913 0.34 0.56 0.92 0.13 0.881925 0.32 0.44 0.78 0.09 0.741929 0.31 0.39 0.71 0.08 0.671933 0.30 0.34 0.65 0.08 0.611938 0.29 0.31 0.61 0.08 0.571950 0.26 0.25 0.52 0.07 0.481955 0.24 0.21 0.46 0.07 0.411960 0.23 0.20 0.43 0.07 0.361965 0.20 0.18 0.38 0.08 0.291970 0.19 0.17 0.36 0.09 0.261975 0.18 0.16 0.34 0.10 0.241980 0.17 0.13 0.31 0.10 0.211985 0.15 0.11 0.26 0.10 0.181990 0.13 0.10 0.23 0.08 0.161995 0.12 0.09 0.21 0.07 0.152000 0.11 0.08 0.19 0.05 0.132005 0.10 0.08 0.18 0.04 0.122010 0.09 0.08 0.17 0.04 0.112015 0.08 0.07 0.15 0.02 0.10
44
Table15
InternationalInequalityinLifeExpectancy,1870-2015(KakwaniIndices)
Unweighted
Populationweighted
MLD Theil Gini Atkinson(ε=2) MLD Theil Gini Atkinson(ε=2)1870 0.21 0.24 0.35 0.30 0.22 0.25 0.36 0.301880 0.23 0.25 0.37 0.32 0.24 0.26 0.37 0.331890 0.26 0.27 0.39 0.36 0.30 0.32 0.42 0.401900 0.28 0.28 0.41 0.38 0.35 0.35 0.44 0.451913 0.27 0.27 0.40 0.38 0.39 0.37 0.46 0.491925 0.30 0.29 0.42 0.41 0.41 0.37 0.47 0.521929 0.26 0.26 0.39 0.38 0.24 0.25 0.38 0.361933 0.25 0.24 0.38 0.37 0.23 0.23 0.37 0.341938 0.21 0.21 0.36 0.32 0.20 0.21 0.35 0.301950 0.16 0.15 0.31 0.28 0.16 0.16 0.31 0.271955 0.15 0.14 0.30 0.27 0.15 0.15 0.30 0.251960 0.15 0.14 0.30 0.26 0.20 0.19 0.34 0.321965 0.13 0.12 0.28 0.24 0.10 0.10 0.25 0.191970 0.12 0.11 0.26 0.22 0.08 0.08 0.22 0.161975 0.11 0.10 0.25 0.22 0.08 0.07 0.21 0.151980 0.10 0.09 0.23 0.19 0.07 0.06 0.19 0.131985 0.09 0.08 0.22 0.18 0.06 0.06 0.19 0.121990 0.09 0.08 0.22 0.19 0.06 0.06 0.19 0.131995 0.10 0.09 0.23 0.21 0.07 0.06 0.19 0.142000 0.11 0.09 0.24 0.21 0.07 0.07 0.20 0.142005 0.11 0.10 0.25 0.22 0.08 0.07 0.21 0.152010 0.11 0.10 0.25 0.21 0.08 0.07 0.21 0.152015 0.09 0.09 0.24 0.17 0.06 0.06 0.19 0.12
45
Table16
InternationalInequalityinLifeExpectancy,1870-2015:TheContributionofChinaandIndiaandSubSaharanAfrica(MLD)(KakwaniIndices)
Excluding Excluding
ActualChinaandIndiaSubSaharanAfrica
1870 0.22 0.20 0.231880 0.24 0.21 0.241890 0.30 0.25 0.311900 0.35 0.26 0.361913 0.39 0.25 0.391925 0.41 0.24 0.411929 0.24 0.23 0.241933 0.23 0.21 0.221938 0.20 0.19 0.201950 0.16 0.16 0.151955 0.15 0.14 0.141960 0.20 0.14 0.201965 0.10 0.12 0.091970 0.08 0.10 0.071975 0.08 0.09 0.061980 0.07 0.08 0.051985 0.06 0.08 0.041990 0.06 0.08 0.041995 0.07 0.09 0.042000 0.07 0.10 0.042005 0.08 0.11 0.052010 0.08 0.11 0.052015 0.06 0.09 0.04
46
Table17
PopulationGrowthDispersion:ContributiontoLifeExpectancyInequality(population-weightedMLDandTheil)(Kakwaniindices)
MLD
Theil
Actual FixedInitialPopulation Actual FixedInitialPopulation1870 0.22 0.22 0.24 0.241880 0.24 0.24 0.26 0.261890 0.30 0.30 0.31 0.321900 0.35 0.35 0.35 0.351913 0.39 0.40 0.37 0.39
1913 0.39 0.39 0.37 0.371925 0.41 0.41 0.37 0.371929 0.25 0.25 0.24 0.241933 0.23 0.23 0.23 0.231938 0.20 0.20 0.20 0.201950 0.17 0.16 0.17 0.16
1950 0.17 0.17 0.17 0.171955 0.15 0.16 0.15 0.151960 0.20 0.21 0.19 0.191965 0.11 0.11 0.10 0.101970 0.09 0.09 0.08 0.081975 0.08 0.08 0.07 0.071980 0.07 0.07 0.06 0.061985 0.06 0.06 0.06 0.061990 0.06 0.06 0.06 0.06
1990 0.06 0.06 0.06 0.061995 0.07 0.07 0.06 0.062000 0.07 0.07 0.07 0.062005 0.08 0.07 0.07 0.072007 0.08 0.07 0.07 0.072010 0.08 0.07 0.07 0.072015 0.06 0.06 0.06 0.06
47
Table18
DecomposingInternationalInequalityinLifeExpectancy(population-weighted):MLDandTheil(KakwaniIndices)
MLD
Within-group Between-group TOTAL OECD TheRest
1870 0.07 0.15 0.22 0.04 0.081880 0.08 0.15 0.24 0.04 0.091890 0.10 0.20 0.30 0.03 0.121900 0.14 0.21 0.35 0.02 0.161913 0.16 0.22 0.39 0.02 0.201925 0.20 0.21 0.41 0.02 0.241929 0.09 0.15 0.24 0.01 0.111933 0.09 0.14 0.23 0.01 0.101938 0.07 0.13 0.20 0.01 0.091950 0.10 0.07 0.16 0.01 0.121955 0.09 0.06 0.15 0.00 0.111960 0.13 0.07 0.20 0.00 0.161965 0.07 0.03 0.10 0.00 0.081970 0.06 0.02 0.08 0.00 0.071975 0.06 0.02 0.08 0.00 0.061980 0.05 0.02 0.07 0.00 0.051985 0.04 0.02 0.06 0.00 0.051990 0.04 0.02 0.06 0.00 0.051995 0.05 0.02 0.07 0.00 0.052000 0.05 0.02 0.07 0.00 0.062005 0.06 0.02 0.08 0.00 0.062010 0.06 0.02 0.08 0.00 0.062015 0.05 0.02 0.06 0.01 0.05
Theil
Within-group Between-group TOTAL OECD TheRest
1870 0.07 0.17 0.25 0.04 0.091880 0.08 0.18 0.26 0.03 0.111890 0.09 0.23 0.32 0.03 0.151900 0.10 0.25 0.35 0.02 0.191913 0.11 0.25 0.37 0.02 0.231925 0.13 0.24 0.37 0.01 0.251929 0.07 0.17 0.25 0.01 0.121933 0.07 0.16 0.23 0.01 0.111938 0.06 0.15 0.21 0.01 0.101950 0.08 0.08 0.16 0.01 0.131955 0.08 0.07 0.15 0.00 0.121960 0.11 0.08 0.19 0.00 0.171965 0.06 0.04 0.10 0.00 0.081970 0.05 0.03 0.08 0.00 0.071975 0.05 0.02 0.07 0.00 0.061980 0.04 0.02 0.06 0.00 0.051985 0.04 0.02 0.06 0.00 0.051990 0.04 0.02 0.06 0.00 0.051995 0.04 0.02 0.06 0.00 0.052000 0.05 0.02 0.07 0.00 0.062005 0.05 0.02 0.07 0.00 0.062010 0.05 0.02 0.07 0.00 0.062015 0.04 0.02 0.06 0.01 0.05
48
Table19
InternationalInequalityinHumanDevelopment,1870-2015
Unweighted
Populationweighted
MLD Theil Gini Atkinson(ε=2) MLD Theil Gini Atkinson(ε=2)1870 0.47 0.43 0.49 0.56 0.53 0.53 0.54 0.581880 0.48 0.43 0.49 0.58 0.55 0.53 0.55 0.611890 0.48 0.42 0.49 0.59 0.59 0.54 0.56 0.641900 0.49 0.40 0.48 0.61 0.62 0.54 0.56 0.681913 0.38 0.32 0.43 0.54 0.51 0.45 0.52 0.621925 0.32 0.27 0.41 0.49 0.45 0.39 0.49 0.581929 0.29 0.25 0.39 0.45 0.33 0.31 0.44 0.461933 0.27 0.24 0.38 0.43 0.29 0.28 0.41 0.421938 0.25 0.22 0.37 0.40 0.27 0.26 0.40 0.401950 0.21 0.18 0.34 0.36 0.23 0.22 0.37 0.351955 0.20 0.17 0.33 0.34 0.20 0.19 0.34 0.311960 0.19 0.16 0.32 0.33 0.20 0.20 0.35 0.321965 0.17 0.15 0.31 0.31 0.15 0.15 0.31 0.261970 0.16 0.14 0.30 0.29 0.14 0.13 0.29 0.231975 0.15 0.13 0.29 0.28 0.12 0.12 0.28 0.221980 0.13 0.12 0.27 0.26 0.11 0.11 0.26 0.191985 0.12 0.10 0.25 0.24 0.10 0.09 0.24 0.171990 0.12 0.10 0.25 0.23 0.09 0.09 0.23 0.161995 0.11 0.10 0.24 0.23 0.08 0.08 0.22 0.152000 0.11 0.10 0.24 0.22 0.08 0.08 0.22 0.152005 0.11 0.09 0.24 0.21 0.08 0.07 0.21 0.142010 0.10 0.09 0.24 0.20 0.07 0.07 0.21 0.142015 0.09 0.08 0.23 0.18 0.06 0.06 0.19 0.11
49
Table20
PopulationGrowthDispersion:ContributiontoHumanDevelopmentInequality(population-weightedMLDandTheil)(Kakwaniindices)
MLD
Theil
Actual
FixedInitialPopulation Actual
FixedInitialPopulation
1870 0.53 0.53 0.52 0.521880 0.55 0.55 0.52 0.521890 0.59 0.58 0.53 0.541900 0.61 0.61 0.53 0.541913 0.51 0.51 0.44 0.46
1913 0.51 0.51 0.44 0.441925 0.44 0.44 0.38 0.381929 0.33 0.33 0.31 0.301933 0.29 0.29 0.27 0.271938 0.27 0.27 0.26 0.261950 0.23 0.23 0.22 0.22
1950 0.23 0.23 0.22 0.221955 0.20 0.20 0.20 0.201960 0.20 0.21 0.20 0.201965 0.15 0.16 0.15 0.151970 0.14 0.14 0.13 0.131975 0.12 0.13 0.12 0.121980 0.11 0.11 0.11 0.111985 0.10 0.10 0.10 0.101990 0.09 0.09 0.09 0.09
1990 0.09 0.09 0.09 0.091995 0.08 0.08 0.08 0.082000 0.08 0.08 0.08 0.072005 0.08 0.07 0.07 0.072010 0.07 0.07 0.07 0.062015 0.06 0.06 0.06 0.06
50
Table21
InternationalInequalityinHumanDevelopment,1870-2015:TheContributionofChinaandIndiaandSubSaharanAfrica(MLD)
Actual
ExcludingChinaandIndia
ExcludingSubSaharanAfrica
1870 0.53 0.41 0.541880 0.55 0.41 0.561890 0.59 0.43 0.601900 0.62 0.42 0.621913 0.51 0.31 0.521925 0.45 0.26 0.451929 0.33 0.24 0.321933 0.29 0.22 0.281938 0.27 0.21 0.271950 0.23 0.19 0.231955 0.20 0.18 0.191960 0.20 0.17 0.201965 0.15 0.16 0.141970 0.14 0.14 0.131975 0.12 0.14 0.111980 0.11 0.12 0.101985 0.10 0.11 0.081990 0.09 0.11 0.081995 0.08 0.11 0.072000 0.08 0.11 0.062005 0.08 0.10 0.062010 0.07 0.10 0.052015 0.06 0.09 0.04
51
Table22
DecomposingInternationalInequalityinHumanDevelopment(population-weighted):MLDandTheil
MLD
Within-group Between-group TOTAL OECD TheRest1870 0.21 0.31 0.53 0.05 0.251880 0.24 0.31 0.55 0.04 0.281890 0.27 0.31 0.59 0.04 0.321900 0.30 0.31 0.62 0.04 0.361913 0.26 0.25 0.51 0.03 0.321925 0.24 0.20 0.45 0.03 0.291929 0.16 0.17 0.33 0.02 0.191933 0.15 0.14 0.29 0.02 0.181938 0.14 0.13 0.27 0.03 0.161950 0.13 0.10 0.23 0.02 0.161955 0.12 0.08 0.20 0.02 0.141960 0.12 0.08 0.20 0.01 0.151965 0.09 0.06 0.15 0.01 0.111970 0.08 0.05 0.14 0.01 0.101975 0.08 0.05 0.12 0.02 0.091980 0.07 0.04 0.11 0.02 0.081985 0.06 0.03 0.10 0.01 0.071990 0.05 0.03 0.09 0.01 0.061995 0.05 0.03 0.08 0.01 0.062000 0.05 0.03 0.08 0.01 0.052005 0.05 0.03 0.08 0.01 0.052010 0.05 0.03 0.07 0.00 0.052015 0.04 0.02 0.06 0.00 0.04
Theil
Within-group Between-group TOTAL OECD TheRest
1870 0.16 0.36 0.53 0.05 0.311880 0.16 0.35 0.53 0.04 0.341890 0.17 0.36 0.54 0.04 0.361900 0.17 0.35 0.54 0.04 0.371913 0.15 0.28 0.45 0.03 0.311925 0.15 0.23 0.39 0.02 0.281929 0.11 0.19 0.31 0.02 0.201933 0.11 0.17 0.28 0.02 0.181938 0.11 0.15 0.26 0.02 0.171950 0.11 0.11 0.22 0.02 0.171955 0.10 0.10 0.19 0.02 0.151960 0.10 0.09 0.20 0.01 0.151965 0.08 0.07 0.15 0.01 0.111970 0.07 0.06 0.13 0.01 0.101975 0.07 0.05 0.12 0.01 0.091980 0.06 0.05 0.11 0.01 0.081985 0.05 0.04 0.09 0.01 0.071990 0.05 0.04 0.09 0.01 0.061995 0.05 0.04 0.08 0.01 0.062000 0.04 0.03 0.08 0.01 0.052005 0.04 0.03 0.07 0.01 0.052010 0.04 0.03 0.07 0.00 0.052015 0.03 0.03 0.06 0.00 0.04
52
Table23
Dimensions’ContributionstoHumanDevelopmentInequality
(population-weighted):MLDandTheilMLD
LifeExpectancy YearsofSchooling AdjustedIncome Residual1870 0.07 0.27 0.01 0.181880 0.08 0.44 0.01 0.021890 0.10 0.46 0.01 0.021900 0.12 0.36 0.02 0.121913 0.13 0.46 0.02 -0.101925 0.14 0.39 0.02 -0.101929 0.08 0.34 0.02 -0.111933 0.08 0.29 0.02 -0.091938 0.07 0.26 0.02 -0.081950 0.05 0.21 0.03 -0.061955 0.05 0.18 0.02 -0.051960 0.07 0.16 0.02 -0.051965 0.03 0.14 0.02 -0.041970 0.03 0.13 0.02 -0.041975 0.02 0.12 0.02 -0.041980 0.02 0.11 0.02 -0.041985 0.02 0.09 0.02 -0.031990 0.02 0.08 0.02 -0.021995 0.02 0.07 0.01 -0.022000 0.02 0.06 0.01 -0.022005 0.03 0.06 0.01 -0.022010 0.03 0.06 0.01 -0.022015 0.02 0.05 0.01 -0.02
Theil
LifeExpectancy YearsofSchooling AdjustedIncome Residual1870 0.08 0.19 0.01 0.251880 0.09 0.30 0.01 0.131890 0.11 0.29 0.01 0.131900 0.12 0.26 0.02 0.141913 0.12 0.31 0.02 0.001925 0.12 0.26 0.02 -0.021929 0.08 0.24 0.02 -0.031933 0.08 0.22 0.02 -0.031938 0.07 0.20 0.02 -0.031950 0.05 0.17 0.03 -0.031955 0.05 0.15 0.02 -0.031960 0.06 0.14 0.02 -0.031965 0.03 0.13 0.02 -0.031970 0.03 0.12 0.02 -0.031975 0.02 0.11 0.02 -0.031980 0.02 0.10 0.02 -0.031985 0.02 0.09 0.02 -0.031990 0.02 0.08 0.02 -0.021995 0.02 0.07 0.01 -0.022000 0.02 0.06 0.01 -0.022005 0.02 0.06 0.01 -0.022010 0.02 0.06 0.01 -0.022015 0.02 0.05 0.01 -0.02
53
Figure1a.UnweightedInequalityinRealPerCapitaGDP,1870-2015
Figure1b.Population-weightedInequalityinRealPerCapitaGDP,1870-2015
54
Figure2.Population-weightedInequalityinRealGDPperHead:TheContributionofChinaandIndiaandSubSaharanAfrica,1870-2015(MLD)
55
Figure3a.DecomposingPopulation-weightedInequalityinRealPerCapitaGDP,1870-2015(MLD).
Figure3b.DecomposingPopulation-weightedInequalityinRealPerCapitaGDP,1870-2015(Theil).
56
Figure4a.UnweightedInequalityinLiteracy,1870-2015(KakwaniIndex)
Figure4b.Population-weightedInequalityinLiteracy,1870-2015(KakwaniIndex)
57
Figure5a.UnweightedInequalityinEnrolment,1870-2015(KakwaniIndex)
Figure5b.PopulationweightedInequalityinEnrolment,1870-2015(KakwaniIndex)
58
Figure6a.UnweightedInequalityinYearsofSchooling,1870-2015(KakwaniIndex)
Figure6b.Population-weightedInequalityinYearsofSchooling1870-2015(KakwaniIndex)
59
Figure7a.Population-weightedInequalityinLiteracy:TheContributionofChinaandIndiaandSubSaharanAfrica,1870-2015(KakwaniIndex)(MLD)
Figure7b.Population-weightedInequalityinEnrolment:TheContributionofChinaandIndiaandSubSaharanAfrica,1870-2015(KakwaniIndex)(MLD)
Figure7c.Population-weightedInequalityinYearsofSchooling:TheContributionofChinaandIndiaandSubSaharanAfrica,1870-2015(KakwaniIndex)(MLD)
60
Figure8a.Population-weightedInequalityinLiteracy(Gini):KakwaniIndexandOriginalValues
Figure8b.Population-weightedInequalityinEnrolment(Gini):KakwaniIndexandOriginalValues
Figure8c.Population-weightedInequalityinSchoolingYears(Gini):KakwaniIndexandOriginalValues
61
Figure9a.DecomposingPopulation-weightedInequalityinLiteracy,1870-2015(KakwaniIndex)(MLD)
Figure9b.DecomposingPopulation-weightedInequalityinLiteracy,1870-2015(KakwaniIndex)(Theil)
62
Figure10a.DecomposingPopulation-weightedInequalityinEnrolment,1870-2015(KakwaniIndex)(MLD).
Figure10b.DecomposingPopulation-weightedInequalityinEnrolment,1870-2015(KakwaniIndex)(Theil)
63
Figure11a.DecomposingPopulation-weightedInequalityinYearsofSchooling,1870-2015(KakwaniIndex)(MLD)
Figure11b.DecomposingPopulation-weightedInequalityinYearsofSchooling,1870-2015((KakwaniIndex)(Theil)
64
Figure12a.UnweightedInequalityinLifeExpectancy,1870-2015(KakwaniIndex)
Figure12b.Population-weightedInequalityinLifeExpectancy,1870-2015(KakwaniIndex)
65
Figure13.Population-weightedinLifeExpectancy:TheContributionofChinaandIndiaandSubSaharanAfrica,1870-2015(KakwaniIndex)(MLD)
Figure14.Population-weightedInequalityinLifeExpectancy(Gini):KakwaniIndexandOriginalValues
66
Figure15a.DecomposingPopulation-weightedInequalityinLifeExpectancy,1870-2015(KakwaniIndex)(MLD)
Figure15b.DecomposingPopulation-weightedInequalityinLifeExpectancy,1870-2015(KakwaniIndex)(Theil)
67
Figure16a.Lifeexpectancy:Population-weightedInequality(Gini)versusLevel(Kakwaniindices)
Figure16bLifeExpectancyPopulation-weightedInequality(Gini)versusLevel(originalvalues)
68
Figure17a.UnweightedInequalityinHumanDevelopment1870-2015
Figure17b.Population-weightedInequalityinHumanDevelopment,1870-2015
69
Figure18.Population-weightedInequalityinHumanDevelopment:TheContributionofChinaandIndiaandSubSaharanAfrica,1870-2015(MLD)
70
Figure19a.DecomposingPopulation-weightedInequalityinHumanDevelopment,1870-2015(MLD)
Figure19b.DecomposingPopulation-weightedInequalityinHumanDevelopment,1870-2015(Theil)
71
Figure20a.Dimensions’ContributionstoHumanDevelopmentInequality(population-weighted):MLD
Figure20b.Dimensions’ContributionstoHumanDevelopmentInequality(population-weighted):Theil
72
Figure21.Population-weightedInequalityinHumanDevelopmentandRealPerCapitaGDP,1870-2015(MLD)
European
Historical
Economics
Society
EHES Working Paper Series
Recent EHES Working Papers
2018
EHES 130
EHES 129
EHES 128
EHES 127
EHES 126
EHES 125
EHES 124
EHES 123
The Napoleonic Wars: A Watershed in Spanish History?
Leandro Prados de la Escosura and Carlos Santiago-Caballero
A New Indicator for Describing Bull and Bear Markets
German Forero-Laverde
The long run impact of foreign direct investment, exports, imports and GDP:
evidence for Spain from an ARDL approach
Verónica Cañal-Fernández and Julio Tascón Fernández
Can Autocracy Promote Literacy? Evidence from a Cultural Alignment Success
Story
Nuno Palma and Jaime Reis
The Origins of the (Cooperative) Species: Raiffeisen Banking in the Netherlands,
1898–1909
Christopher L. Colvin, Stuart Henderson and John D. Turner
‘Getting to Denmark’: the Role of Elites for Development
Peter Sandholt Jensen, Markus Lampe, Paul Sharp and Christian Volmar
Skovsgaard
How to become a leader in an emerging new global market: The determinants of
French wine exports, 1848-1938
María Isabel Ayuda, Hugo Ferrer-Pérez and Vicente Pinilla
Still a long way to go: decomposing income inequality across Italy’s regions,
1871 – 2011
Gabriele Cappelli, Emanuele Felice, Julio Martínez-Galarraga and Daniel Tirado
All papers may be downloaded free of charge from: www.ehes.org The European Historical Economics Society is concerned with advancing education in European economic
history through study of European economies and economic history. The society is registered with the
Charity Commissioners of England and Wales number: 1052680