Urban!Biased!Social!Policies!and!the!Urban3Rural!Divide!in!China! by! Kaijie!Chen! Department!of!Political!Science! Duke!University!

Transcription

1 UrbanBiasedSocialPoliciesandtheUrban3RuralDivideinChina by KaijieChen DepartmentofPoliticalScience DukeUniversity Date: Approved: ProfessorKarenRemmer,Supervisor ProfessorPabloBeramendi ProfessorAnirudhKrishna Thesissubmittedinpartialfulfillmentof therequirementsforthedegreeofmasterofartsinthedepartmentof PoliticalScienceintheGraduateSchool ofdukeuniversity 2015

2 ABSTRACT UrbanBiasedSocialPoliciesandtheUrban3RuralDivideinChina by KaijieChen DepartmentofPoliticalScience DukeUniversity Date: Approved: ProfessorKarenRemmer,Supervisor ProfessorPabloBeramendi ProfessorAnirudhKrishna Anabstractofathesissubmittedinpartial fulfillmentoftherequirementsforthedegree ofmasterofartsinthedepartmentof PoliticalScienceintheGraduateSchoolof DukeUniversity 2015

3 Copyrightby KaijieChen 2015

4 Abstract Thisthesisprovidesacasestudyoftheurban3ruraldivideanditsrelationshipto public polices in China. By utilizing multiple data sources from both government statistics and independent surveys, this study examines the trends of China s urban3 rural divide and analyzes the influences of three public policies: 1) the household registrationsystem,2)theeducationsystem,and3)thehealthcaresystemonthedivide. The findings show that urban3rural gaps in both economic and social dimensions are largeandhavebeenincreasingsincechina smarketreformofthelate1970s.thegaps are closely associated with the strong urban bias imbedded in the public policies. Specifically, the household registration system differentiates the urban and rural populations, provides the urban population with privileges in income and education, and restricts the rural population s access to urban resources. The education and healthcare systems further strengthen the gaps through their dualistic structures and differentfundingschemesimplementedinurbanandruralareas. iv

5 Contents Abstract...iv ListofTables...vi ListofFigures...viii 1.Introduction StudyingtheUrban3RuralInequalityinChina TheUrban3RuralDivideinChina IncomeGap EducationalGap InequalityinHealthcareandInfrastructure TheUrban3RuralDivideattheProvincialLevel OverallTrendsintheUrban3RuralGap SocialPoliciesandtheUrban3RuralGap HouseholdRegistrationSystem EducationSystem HealthcareSystem DiscussionandConclusion...66 References...71 v

6 List%of%Tables Table1:PerCapitaIncomeofUrbanandRuralHouseholdinChina, (China StatisticalYearbook,2014)...13 Table2:IlliteracyRate,Urbanvs.Rural,2000,2010.(Census,2000,2010)...17 Table3:PerCapitaHealthcareExpenditureoftheUrbanandRuralHousehold,1990, 1995,2000, (NBS,2014)...23 Table4:Urban/RuralRatioinPerCapitaHouseholdIncomebyProvince,2000,2005, 2010,2013.(NBS,2014)...26 Table5:UrbanandRuralPerCapitaHouseholdIncomeandUrbanizationinShanghai, (ShanghaiStatisticsBureau,2014)...28 Table6:UrbanandRuralPerCapitaHouseholdIncomeandUrbanizationinYunnan, (YunnanStatisticsBureau,2011)...29 Table7:IndividualTotalIncomebyHukouStatus,1995,2002.(CHIP,1995,2002)...39 Table8:Income,YearsofEducation,andHukouStatusoftheTopandBottomIncome Quintiles,1995,2002.(CHIP,1995,2002;Yueetal,2008)...40 Table9:EducationLevelbyHukouStatus,1995,2002,2005.(CHIP,1995,2002;Wu, 2012)...42 Table10:MarginalReturnstoEducationatMeanEducationLevels,Urban,Rural,1995, 2002.(yuan)(Yueetal,2008)...43 Table11:ImportantEducationPoliciesinChina, (TheCentralPeople s GovernmentofChina,2013)...1 Table12:PercentageofNationalEducationBudgetDistributionbyEducationLeveland Sector, (%)(MinistryofEducation,2006,2007,2008,2009,2010,2011)...53 Table13:EducationalAttainmentofPrimarySchoolandJuniorHighSchoolTeachers, Urbanvs.Rural,2010.(MinistryofEducation,2010)...55 Table14:ProfessionalRankofPrimarySchoolandJuniorHighSchoolTeachers,Urban vs.rural,2010.(ministryofeducation,2010)...55 vi

7 Table15:GrossPrimarySchoolandJuniorHighSchoolEnrollmentRatio,Urbanvs. Rural,1990,2000,2005,2010.(MinistryofEducation, ;Chen(2013))...57 Table16:PublicHealthcareProgramsinChina...59 Table17:PercentageofPopulationCoveredbyPublicHealthcarePrograms,Urbanvs. Rural,2005, (NBS,2014;NationalHealthandFamilyPlanningCommission, 2014)...62 Table18:AveragePerCapitaSubsidiesfromDifferentHealthcarePrograms,2005, (NationalHealthandFamilyPlanningCommission,2013)...63 vii

8 List%of%Figures Figure1:ChangeofUrban/RuralHouseholdPerCapitaIncomeRatio, Figure2:AverageYearsofEducationofPopulationover15YearsOld,National,Urban, Rural,1988.(Gustafsson,Shi&Sicular,2008)...16 Figure3:Urban3RuralDifferenceinAverageYearsofEducation, (Li,2006).18 Figure4:Urban3RuralDifferenceinPerCapitaEducationalExpenditure, (Li, 2006)...19 Figure5:NumberofHospitalBedsper1000People,Urbanvs.Rural,1978,1980,1983, (NBS,2000)...21 Figure6:Number of Healthcare Personnel per 1000 People, Urban vs. Rural, 1978, 1980, 1983, (NBS, 2000)...22 Figure7:AdministrationStructureofChina...48 viii

9 1.#Introduction Over the past several decades, the market economy and globalization have arguably been the two most important influences on the economy of developing countries. Most of the poor and developing countries have either voluntarily or involuntarilyadoptedmarketreform,whichwasbelievedtonotonlypromoteeconomic developmentbutalsohelpredistributionandpovertyreduction(bauer,1984).contrary tothepredictionsofliberaleconomics,however,manydevelopingcountrieshavebeen experiencing increasing inequality despite their economic achievement. As one of the most dynamic developing countries in the world, China adopted unilateral market reforminthelate1970sandhasachievedunprecedentedrapideconomicgrowthover thepastfourdecades.however,thechineseeconomicmiraclehasbeencharacterized byincreasingandpersistinghighinequality,whichispredominantlydrivenbyspatial inequalitybetweenurbanandruralareas(yang,1999;theworldbank,1997). Today s China is marked by two distinct economies: one urban or modern economywithrapidexpansion,significantforeigninvestmentandtrade,andsomeof theworld smostadvancedindustries,andoneruralorlaggedeconomywithtraditional agricultureandextremepoverty.theurban3ruralinequalityinchinahasattractedthe attentionfromscholarsacrossvariousdisciplines,buttheavailableliteratureislimited intwokeyrespects.first,mostofthestudiesfocusexclusivelyoneconomicinequality. 1

10 Second, market mechanism and globalization have usually been utilized as the independent variables. More recent studies started to question the role of political institutions, public policies, and power relations in shaping the urban3rural divide in China.Mostnotably,thestudyofurban3biasedpoliciesandspatialinequalityhasbeen appliedtothecaseofchina. Inspiredbytheearliestworksonurbanbiasandtherecenteffortsinapplying theconcepttochina,thispaperattemptstoprovideadetailedinvestigationofwhether urban3biasedsocialpolicieshavecontributedtotheincreasingurban3ruraldivide.there aretwoprimaryaimsofthepaper:thefirstistoexaminethedevelopmentoftheurban3 rural inequality in China, including economic and social inequality. The second is to analyze how social policies have influenced the urban3rural inequality with specific focusonthreesetsofpolicies:thehouseholdregistrationsystem(hukou),theeducation system,andthehealthcarepolicy. ThispapercontributestotheongoingstudyonChina surban3ruralinequalityin three ways. First, it expands the study of inequality beyond economic inequality to inequalityinothersocialdimensions.second,socialpoliciesandpoliticalinstitutionsare theprimaryindependentvariablesratherthanmarketmechanism.third,thispaperalso joinstherecenteffortsinunderstandingthemechanismoftheinequalityindeveloping 2

11 countries through detailed micro3level and case study to explore the relevance of China sexperienceforotherdevelopingcountries. Thepaperisdividedintofoursections.Thefirstsectionprovidesthebackground of the study and discusses how previous studies have approached the question of urban3ruralinequalityinchina.thesecondsectiondrawsapictureoftheurban3rural inequality in both economic and social dimensions by examining multiple resources from both government data and independent studies. The third section provides an analysis of the influences of social policies and government decision3making on the increasingurban3ruraldivide.thelastsectionsummarizesthekeyconclusionsthatmay bedrawnfromtheanalysisanddiscussesimplicationsforfutureresearch. 3

12 2.#Studying#the$Urban(Rural&Inequality&in&China& China started unilateral market reform in the late 1970s and conducted rapid marketreformoverthe1980sand1990s.in2001,chinaofficiallyjoinedtheworldtrade Organization(WTO)andbecameoneoftheleadingtradepartnerstobothdeveloped anddevelopingcountries.chinahasoneofthedevelopingworld smostopentradeand financial regimes, and scholars have argued that China has become the most open economy among developing countries (Branstetter & Lardy, 2006). Starting from its market liberalization, China s economy has skyrocketed. Over the past two decades, China has maintained its annual GDP growth rate over 8 percent (The World Bank, 2015),oneofthehighestintheworld.In1978,theyearwhenChinastarteditsreform andopen3uppolicy,china sgdpwas$148billion,butthisnumberincreasedto$5.93 trillionin2010,whenchinasurpassedjapantobecomethesecondlargesteconomyin theworld.by2014,china sgdpreached$9.24trillion,whichisover60timesmorethan thatin1978(theworldbank,2015;nationalbureauofstatisticsofchina(nbs),2014). Economicgrowthhasalsoresultedinrapidurbanization.In1978,onlyaround 18percentofChinesepopulation(0.17billion)livedinurbanareas,butthisnumberhas beensteadilyincreasingwithanaverageannualurbanpopulationgrowthratearound4 percent.overhalfofchinesepopulation(53percent)liveinurbanareastoday(world Bank, 2015). At the same time, we have seen the decrease of rural population as a 4

13 percentageoftotalpopulation.before1991,ruralpopulationhadbeengrowingslowly (around0.5percentannually),mostlyduetothenaturalpopulationgrowth;butstarting from 1991, the process of urbanization has counteracted the effect of the natural populationgrowthinruralareasandtheruralpopulationgrowthratehasbeennegative. However,whatisnoteworthyisthatasubstantialportionofChinesepopulation(0.636 billion, or 47 percent of the whole population) still lives in rural areas, where the majorityofchinesepoorpopulationisconcentrated. Despite its economic success over the past several decades, China s overall inequality,measuredbyginicoefficient,hasbeenreportedtoincreaserapidly.thereare disagreementsovertheexactginicoefficientofchinaduetotheproblemsofchina s government statistics, which have been criticized for their lack of credibility and transparency (see Hvistendahl, 2013). In 2000, China s Gini coefficient was 0.41 accordingtogovernmentstatistics(nbs,2014),butthepublicationofthismeasurehad been stopped since 2000 for unknown reasons and this makes it harder to trace the change of the overall inequality in China over the past decade. In 2013, the Chinese governmentpublishedthedataagain,andginicoefficientwasreportedtobeslightly lower than 0.5. Due to the ambiguity of government data, other scholars and third parties have estimated China s Gini coefficient from other resource. Khan and Riskin (1998)foundthatChina sginicoefficienthadincreasedfrom0.38in1988to0.45in

14 AccordingtotheestimateofTheWorldBank(2015),China sginicoefficientin1981was 0.29,butthenumberincreaseddramaticallyduring1980sand1990sandreached0.43in 2002.TheoverallinequalityinChinakeptatthesamehighlevelforthelastdecade. A more recent study (Xie & Zhou, 2014) estimated that the Gini coefficient in Chinahasincreasedtoamuchhigherlevelbetween0.53and0.55since2005.Moreover, theauthorsfoundthatchinahasbeenexperiencingamuchfasterincreaseininequality leveloverthepastdecadethanbefore.althoughbothofficialstatisticsandstudiesdone byindependentresearcherscannotprovideadecisiveanswertohowunequalischina today,oneimportanttrendworthnotinghereisthattheinequalitylevelinchinahas been increasing along with its market reform and embrace of globalization. In comparison with other developing countries (for example, India, Indonesia, and Vietnam s Gini coefficients are 0.33, 0.43, and 0.36, respectively) and other Asian countries (for example, Japan s Gini coefficient is 0.38), China has become one of the mostunequaleconomiesamongdevelopingcountriesandinitsownregion. Studies have found that China s inequality is predominately driven by spatial inequality,especiallytheurban3ruraldivide.xieandzhou(2014)foundthattheurban3 ruraldivideandtheprovincialinequalityarethetwomostimportantforcesunderlying China s overall inequality. One report from The World Bank (1997) argued that the urban3ruralincomeinequalityexplainedmorethanhalfoftheoverallincomeinequality 6

15 in China in 1995, and also explained more than 75 percent of the change in overall inequality in 1980s and 1990s. Another study in Jiangsu and Sichuan provincesfound thattheurban3ruraldividealmostexplainedallofthechangeinoverallinequalityfrom 1986to1994inthesetworegions(Yang,1999). Given the high level of inequality in China and the prominence of the urban3 ruraldivide,scholarshavetriedtoexaminewhetherthemarketliberalizationandthe rapid structural adjustment are the reasons behind. Classic studies on market reform andinequalityusecross3countryregressionsandhaveproducedmixedresults.forbes (2000)foundthateconomicgrowthcanleadtorisinginequality,butDollarandKraay (2002)arguedthatgrowthandthemarketeconomyareactuallygoodforredistribution anddonotnecessarilyleadtoinequality. Cross3countrystudieshavebeencriticizedbyotherscholarswhoarguethatthe datafromdevelopingcountriesareusuallyoflowqualityandsignificantvariationsof politicalinstitutionsandmeasuresmakecomparisonsofinequalityamongdeveloping countriesproblematic.(ravallion,2001).oneclassicalstudybygoldbergandpavcnik (2007)gatheredmostresearchevidenceontheeffectsofglobalizationoninequalityand found that no measure of inequality in developing countries has shown a decrease duringtheperiodofrapidmarketliberalizationandglobalization,andmanymeasures haveincreased.however,theyfamouslyarguedthat themechanismsthroughwhich 7

16 globalizationhasinfluencedinequalityarecountry,time,andcasespecific (Goldberg& Pavcnik,2007).Thereislittledoubttodaythatmarketreformandglobalizationdohave stronginfluencesoninequality,althoughtheymightnotbethedirectcauses. Considering the limitations of classical studies and the need for more detailed case studies, China scholars started to examine the specific mechanisms and factors influencingtheurban3ruraldivideinchina.weiandwu(2001)didthefirststudyon tradeopennessandinequalityinchina,andtheyfoundthattheurban3ruralinequality surprisingly declined in regions that experienced greater degree of openness in trade. The failure of trade openness to explain the growing urban3rural divide led other scholars to examine the influence of political institutions and public policies. For example,smokeandkim(2003)reportedthatthebiasedfiscaltransfershaveresultedin inequalityinasiancountries.inthecaseofchina,wong(2007)arguedthatthefiscal transfers in China favor regions with larger tax bases and less than one third of the transfersaredistributedtothepoorregions.thus,thepopulationinpoorareasreceived less transfers from the governmentthanthe populationinmoreeconomicallyprosper areas. Malesky, Abrami, and Zheng (2011) compared the cases of China and Vietnam andconcludedthatthechinesegovernmentdevotesmuchlessbudgettotransfersand redistributionthanvietnam,andtherelativelysmallerdecision3makingbodyinchina 8

17 provideslessincentivesondevelopingstrategiesandpoliciesforequalizingandpoverty reduction. Although previous studies have produced important findings about China s inequality, they provide only partial answers to our understanding of the urban3rural divideinchina.westilldonothaveacomprehensivepictureofthereasonsbehindthe growingandpersistingdivide.asanattempttojointheongoingeffortsinexplaining China surban3ruraldivide,otherscholarshaveintroducedthestudyofurbanbiasand examined its implication to China. The development of urban bias theory should be credited to two important works. In his influential book Why' Poor' People' Stay' Poor, Lipton(1977)arguedthaturban3ruralconflictisthemostimportantclassconflictinthe developingworld,withconcentratedresourcesandpower,theurbanclassesareableto designandimplementdevelopmentplansthatare from,byandforthepeopleincities (p.68).asaresult,decision3makingprocessandpoliciesthatarepredominantlymade byurbanelitesareusuallyignoranttotheneedsoftheruralpopulation. In his study of African countries, Bates (1981) argued that policies that favor urban development control prices of agricultural products, leave the cost of developmenttotheruralpopulation,andlimittheircapabilitytobecomebetteroff.both oftheauthorsfindthaturbanbiasisresultedfromtheconcentrationofresourcesand politicalpowerinurbanareas,wherepoliciesaredevelopedinfavorofurbaneconomic 9

18 development. The key question in applying urban bias theory to the case of China is whether China is also an example of urban bias, if so, what policies and government decision3makingprocesseshavecontributedtosuchbias?asmallbutgrowingliterature inthisareahasalreadyshownsomeevidenceofurbanbiasinchina. Oi(1990,1993)didseveralstudiesonChina sagriculturalpoliciesandfoundthat governmentsubsidiestourbanareasonfoodandhousinghavesignificantlycontributed to the urban3rural income inequality in the early years of market reform. Khan et al (1993) examined urban3biased housing policy and found that housing subsidies accountedforover30percentofurbanincomeinthe1980sandnosuchsubsidieswere available in rural areas. Yang (2007) argued that urban areas have benefited from development plans biased towards heavy3industry development, and the agricultural surplushasbeenextractedforthecapitalaccumulationinurbanareas.althoughmostof the studies on China s urban bias focus on economic policies, a few studies have inquiredintotheroleofsocialpolicies.wuandtreiment(2004)studiedtheroleofthe householdregistrationsystemininfluencingsocialstratificationandincomeandfound that the system favors the urban population who has access to more occupation opportunitieswithhigherpayment.however,mostoftheexistingstudiesofurbanbias in China focus exclusively on economic inequality and economic policies without considerationofinequalityinotherdimensionsandtheroleofsocialpolicies. 10

19 Infillingthisgapandjoiningtheeffortsinunderstandingurban3biasedpolicies in China, this study attempts to first examine the trends of the urban3rural divide in income,education,healthcare,andothersocialdimensions,andthenstudyhowpublic policieshaveinfluencedthesetrends.basedonmyanalysis,ifindthattheurban3rural dividehasbeenincreasinginbotheconomicandsocialdimensionssincethelate1970s whenchinastartedmarketreform.urbanbiasembeddedinthehouseholdregistration system favors the urban population by providing privileges in income and access to educationalresourcesandsocialservices.thesystemalsorestrictstheruralpopulation from accessing to urban resources through strict control on rural3to3urban migration. With the establishment of the urban3rural distinction by the household registration system, the dualistic structures and different funding schemes of education and healthcaresystemsinurbanandruralareasfurtherstrengthenandenlargetheurban3 ruralgap. 11

20 3.#The$Urban(Rural&Divide&in&China The large urban3rural divide in China is not new, but starting from China s policy reforms of the late 1970s, the divide has been increasing and has received significant attention. Moreover, its contribution to overall inequality has also been increasing.in1988,itscontributionwas37percent.thatnumberincreasedto41percent in1995,andreachedanewhighas46percentin2002(gustafsson,shi,&sicular,2008). With the latest data from the NBS and other nationally representative surveys, the urban3rural divide in China is reexamined here to include inequality along both economic and social dimensions, the latter of which have been largely ignored by previousstudies. 3.1IncomeGap Urban3ruralincomedifferenceshavereceivedthemostintensiveattentionfrom scholars. Here I use the most recent government data to show the change in these differencessincechina smarketliberalization.thechina'statistical'yearbook,published annually by NBS, is the only comprehensive source of government statistics on the nationaleconomyandlivingconditionsofthechinesepopulation.table'1presentsthe dataonpercapitaincomeofurbanandruralhouseholds.thedatashowsthatpercapita income of both urban and rural households have been increasing since 1978, the year whenchinaofficiallystartedmarketreform.moreimportantly,theurban/ruralratiohas 12

21 alsobeenincreasing.beginningin2002,percapitaincomeofurbanhouseholdsisover threetimesgreaterthanthatofruralhouseholdsandthistrendcontinuestothepresent. Table1:PerCapitaIncomeofUrbanandRuralHouseholdinChina,1978(2013. (ChinaStatisticalYearbook,2014). Year Per capita income of urban household (yuan) Per capita income of rural household (yuan) Urban-rural ratio The gap is large by international standards; Eastwood and Lipton (2004) have reportedthattheurban3ruralincomeratioofmostasiancountriesfallsbetween1.3and 1.8,withthePhilippinesasanoutlierwitharatioof2.2.Anothersource(Knight&Song, 13

22 1999)studiedtheurban3ruralincomeratioin12countriesinAsia,theMiddlesEast,and Africa, and found that China s ratio is the third highest after Zimbabwe and South Africa.Tofurtherexplorethechangingtrendsoftheurban3ruralhouseholdincomeratio, Figure'1 shows the change of the urban3rural per capita household income ratio from 1978to2013. Figure1:ChangeofUrban/RuralHouseholdPerCapitaIncomeRatio, As can be seen, the urban3rural income gap first decreased significantly in the early years of market reform, probably due to the immediate economic benefits generatedinbothurbanandruralareasandtheblurredurban3ruraldifference.thegap 14

23 thenincreaseddramaticallybetween1985and1995.aftergoingthroughashortperiod of decline in 1996, it increased steadily from 1998 to 2009, by which time urban householdspercapitaincomewasover3.3timesgreaterthanthatofruralhouseholds. In2010,thegapstartedtodeclinegradually.Butingeneral,overthepastthreedecades, theurban3ruralincomegapinchinahasincreasedtoaveryhighlevel. The significant increase and prominence of the urban3rural income gap, especiallyoverthelasttwodecades,issurprisinggiventheequalizingforcesfromthe centralgovernmentandtheincreaseofrural3to3urbanmigrationflows.beginninginthe late 1990s, the Chinese government has developed equalizing strategies to boost the ruraleconomy.thesestrategiesincludereductionofhousingsubsidiesinurbanareas, agriculturaltaxreduction,andinvestmentininfrastructureandeducationinruralareas. Butthegrowinggapbetweenurbanandruralareasoccurreddespitethegovernment s equalizing efforts. Similarly, although labor migration from rural areas to urban areas hasincreasedconsiderablysince1990s,migrationseemstohavehadalimitedimpacton theincreasingurban3ruralgap.khanandriskin(2008)foundthatincludingmigration onlyreducestheincomegapby6percent. 3.2EducationalGap The urban and rural populations are also found to differ significantly in measuresofeducationalattainmentandaccesstoeducationalresources.thedataonthe 15

24 urban3rural differences in educational measurements is fragmented, and no consistent dataispublishedonthesemeasureseveryyear,makingitdifficulttotrackthechanges of the urban3rural educational inequality. Here I use data from a 1988 national householdsurveyconductedbychineseacademyofsocialsciences(cass)reportedin Gustafsson,Shi&Sicular(2008),thecensusdatafor2000and2010,andsourcesfrom independentstudiestodepictapictureofthetrendsintheeducationalgap. Figure2:AverageYearsofEducationofPopulationover15YearsOld, National,Urban,Rural,1988.(Gustafsson,Shi&Sicular,2008). Figure'2showstheaverageyearsofeducationofthepopulationover15yearsold by different sectors in As shown, significant gaps in years of education existed 16

25 betweenurbanandruralareasacrossallagegroups.averageyearsofeducationinrural areas were also significantly lower than the national level. On average, the urban populationhad9.6yearsofeducationin1988,butthenumberfortheruralpopulation wasonly5.5years,resultinginadifferenceof4.1years.sincethecensusdatafor2000 and 2010 does not measure the average years of education of the urban and rural populations, I use instead the data on the illiteracy rate to show the educational gap observedin2000and2010andtable'2presentsthedata.ascanbeseen,acrossallage groups,theilliteracy rateinurbanareasislowerthanthatinruralareasinboth2000 and2010.althoughtheilliteracyrateinbothurbanandruralareaswaslowerin2010 than in 2000, the gaps across all age groups still persist. However, due to the discontinuityandinconsistencyofgovernmentdata,itishardtocomparethelevelof educationoveralongertimeperiodorexaminethechangeoftheeducationalgapover thepastfewdecades. Table2:IlliteracyRate,Urbanvs.Rural,2000,2010.(Census,2000,2010). Age group 2000 Urban (%) 2000 Rural (%) 2010 Urban (%) 2010 Rural (%)

26 Tocomplementthedatapresentedabove,Ipresenthereanindependentstudy donebyli(2006),whoexaminedthechangeoftheurban3ruraleducationalinequality between1990and2005inchina.figure'3andfigure'4presenttheurban3ruraldifference in average years of education and average educational expenditure per capita, respectively. Figure3:Urban(RuralDifferenceinAverageYearsofEducation,1990(2005.(Li, 2006). 18

27 Figure4:Urban(RuralDifferenceinPerCapitaEducationalExpenditure,1990( 2005.(Li,2006). AsshowninFigure'3,theaverageyearsofeducationoftheruralpopulationwas lessthanthatintheurbanpopulationwithadifferenceofover2yearsbetween1990and Although the number fluctuated between 1990 and 2005, we can see that after reachingapeakin1997,thedifferenceinlateryearsislargerthanthatinearlieryears, withtheexceptionof2000.ingeneral,thegapinaverageyearsofeducationhasgrown from 1990 to Figure'4 showstheurban3rural difference in per capita educational expenditure between 1990 and Education expenditure is another common measurementofeducationalinequality,andthetrendsinfigure'4clearlyshowsthatthe difference in educational expenditure between the urban and rural populations 19

28 increasedsignificantlybetween1990and2005,especiallyinmorerecentyears.by2005, thegapineducationalexpenditurereached788.72yuanperyear,whichis9timesmore than that of The significant increase in the education expenditure is associated withtheincreasingincomegappresentedearlier,andthereisnosignthatthegaphas narrowedinmorerecentyears,giventhattheincomegaphasalsobeeincreasingafter Thefindingspresentedherehaveshownthatonlythemeasureontheilliteracy rateshowsadecliningurban3ruralgapovertime.bothaverageyearsofeducationand educationalexpenditureprovidestrongevidenceoftheincreasingeducationalgap.the prominence of the educational gap has also been increasing in recent years, Qian and Smith (2008) show that the urban3rural educational inequality is the most important contributor to the overall growth of educational inequality in China after the coastal3 inlandgap,makingtheginicoefficientofeducationinchinaamongthehighestinthe world. 3.3InequalityinHealthcareandInfrastructure While the income and educational inequality have received some attention in previousstudies,theurban3ruralgapsinotheraspectshavenotbeenwellstudied.one of the most important gaps exists in healthcare. China s healthcare system and its investmentinhealthcarehavebeenwidelyreportedtobeoneoftheworstintheworld 20

29 (Zhang & Kanbur, 2005). According to WTO (2000), China is ranked 144th in overall healthcaresystemperformanceand139thinhealthexpenditurepercapitaworldwide. What is worse is the widening healthcare gap between urban and rural areas. Some scholarsreportedthatchinaisamongthecountrieswiththemostunequalhealthcare financeintheworld(gustafsson,shi&sicular,2008).drawingontwodatasetsfromthe China' Statistical' Yearbook, this section attempts to provide some evidence on the wideningurban3ruralgapinhealthcareresourcesandinvestment. Figure5:NumberofHospitalBedsper1000People,Urbanvs.Rural,1978, 1980,1983,1985(1998.(NBS,2000). 21

30 Figure6:Number of Healthcare Personnel per 1000 People, Urban vs. Rural, 1978, 1980, 1983, (NBS, 2000) Figure'5andFigure'6presentthedatafromNBSonthenumberofhospitalbeds andthenumberofmedicalpersonnelper1000peopleinurbanandruralareasbetween 1978 and As can be seen, the number of hospital beds and the number of healthcarepersonnelper1000peopleinurbanareasweresignificantlyhigherthanthose inruralareas.theurban3ruralgapsinthesetwoindicatorsalsoshowsimilarpatterns overtime:afterfluctuatingbetween1978and1985,bothofthegapsincreasedsteadily from1985to1998.in1985,theurban3ruraldifferencesinthenumberofhospitalbeds andthenumberofmedicalpersonnelper1000peoplewerearound3and1,respectively, 22

31 whilethedifferencesincreasedtoaround5and3.2,respectively,in1998.theenlarging urban3rural gaps in medical facilities and medical personnel are surprising given China srapideconomicdevelopmentanditsincreasedinvestmentinhealthcaresector andhealthcarereforminbothurbanandruralareas. Table3:PerCapitaHealthcareExpenditureoftheUrbanandRural Household,1990,1995,2000,2010(2013.(NBS,2014). Urban (yuan) Rural (yuan) Difference (yuan) Table'3presentsthedataonpercapitahealthcareexpenditureoftheurbanand ruralhouseholdsin1990,1995,2000,andbetween2010and2013.althoughthedatais not available for each single year, it can clearly be seen that the urban3rural gap in healthcareexpenditureincreaseddramaticallyfrom1990to2010.afteradeclinein2011, the gap reached its highest point in 2012 and declined again in The data and evidence presented above have shown support for the widening urban3rural gap in accesstohealthcarefacilitiesandexpenditure. Anotherimportantpartoftheurban3ruralinequality in China canbefoundin infrastructure.althoughthechinesegovernmenthasdevotedalotinconstructionand 23

32 improvementofinfrastructureinruralareas,theurban3ruralgapstillpersists.theonly availabledataontheruralinfrastructureistheagricultural'censusin1996(nbs 1996), the1996dataindicatesthatthedevelopmentofinfrastructureinruralareasisfarbehind thatinurbanareas.althoughmostruralhouseholdshadaccesstowaterandelectricity, almost no household had access to gas. In comparison, the percentage of the urban population with access to gas is 73.2 percent in 1996 (NBS, 1997). In terms of road density,thelengthofroadper10thousandpeopleinurbanandruralareaswas7km and18.4km,respectively,in1996.(nbs,1996,1997).thehighernumberinruralareas shouldbeattributedtothefactthatthesizeofruralareaandruralpopulationislarger thantheurbanareaandtheurbanpopulation.however,mostoftheroadconstruction in rural areas is of low quality, and probably the most important reason is the poor qualityofgovernanceinruralareas. Although no dataset about the quality of governance in rural China has been established,somerecentstudieshavealreadyfoundsomeevidenceonthis.onesurvey studyofqualityofgovernanceinchina(saich,2012)foundthatchinesepeoplethink villagegovernmentsarelesscompetentthancitygovernments,andgivelowerscoresto village governments in terms of public service provision. And rural residents are less satisfiedwiththeirgovernmentsthanurbanresidents.anotherstudyhasfoundthatthe quality of village projects in China is associated with the poor quality of governance; 24

33 communitieswithproblemsingovernmentcapability usually cannotimplementgood projects (Liu et al, 2013). Given that infrastructure is found to promote agricultural productivity and nonfarm employment in the rural areas (Fan & Zhang, 2004), infrastructurecanbecomeanothersourceoftheurban3ruralinequality. 3.4TheUrban(RuralDivideattheProvincialLevel Althoughthedataatthenationallevelhasshownthattheurban3ruraldivideis largeandincreasing,itispossiblethatdifferenttrendscanbeobservedattheregionalor the provincial level. To see whether the urban3rural gap varies across regions within China,theNBSdataontheurbanandruralpercapitahouseholdincomein31provinces isexaminedandtheurban3ruralincomeratiosin2000,2005,2010,ad2013arecalculated andpresentedintable'4.althoughthedatabefore2000isnotavailableinnbs,making ithardtocomparethetrendsattheprovincialleveltothetrendsatthenationallevelin theearlytimeofthemarketreform,severalimportantfindingscanbedrawnhere.first, mostprovinces(28outof31)experiencedanincreaseintheurban3ruralincomegapin 2005incomparisonwith2000.Second,from2005to2010,situationvariedindifferent provinces,butnotmanyprovincesexperiencedasignificantchangeandthegapsstayed roughlyatthesameleveloverthe53yearperiod.third,in2013,mostprovinces(29out of 31) experienced a decline in the gap. These three general trends mirror the observationsfoundatthenationallevel. 25

34 Table4:Urban/RuralRatioinPerCapitaHouseholdIncomebyProvince,2000, 2005,2010,2013.(NBS,2014) National Beijing Tianjin Hebei Shanxi Inner Mongolia Liaoning Jilin Heilongjiang Shanghai Jiangsu Zhejiang Anhui Fujian Jiangxi Shandong Henan Hubei Hunan Guangdong Guangxi Hainan Chongqing Sichuan Guizhou Yunnan Tibet Shaanxi Gansu Qinghai Ningxia Xinjiang Despite the fact that the trends in most provinces are similar to those at the nationallevel,significantvariationsexistamongprovincesintermsofthemagnitudeof the gaps, the urban3rural income gaps in some provinces are significantly larger than 26

35 thoseinotherprovinces.ascanbeseen,provinceswiththelargesturban3ruralincome gaps are Guizhou, Yunnan, Tibet, Shaanxi, and Gansu, all of which are located in the southwestandnorthwest,theinnermostpartsofchina.in2005,allthefiveprovinces hadincomeratiosover4,in2013,whentheincomegapsinmostoftheprovinceswent belowthenationalaverage,thefiveprovincesstillhadsignificantlyhigherincomegaps whencomparingtootherprovinces,especiallymoreeconomicallyprosperareasalong thecoastalregion.onepossibleexplanationfortheextremelylargeurban3ruralincome gapsinsomeprovincesmayberelatedtotherelativesizeofruralareastourbanareas and the different development level. The southwest and northwest of China are the poorest regions in China and the latest to experience market reform and opening3up. Agricultureisthemajoreconomicsectorandasignificantportionofthepopulationstill livesinruralareas.economicallyadvancedareasmayhaverelativesmalloralmostno rural areas due to rapid economic development and urbanization. Also, the developmentlevelofruralareasindifferentregionsmayalsodiffersignificantly.with smaller size of rural areas and relatively higher level of rural development, provinces withbettereconomicdevelopmentmayhavesmallerurban3ruralgaps. To see if the hypothesis is correct and provide further evidence on the urban3 ruraldivideattheprovinciallevel,ipresenttwocasestudieshereonyunnan,whichis oneofthepoorestregionsinchinawiththehighesturban3ruralincomeratioin2013, 27

36 andshanghai,whichistheeconomiccenterofchinawithoneofthelowesturban3rural incomeratiobetween2000and2013.ilookatthedataontheurbanandruralpercapita householdincomeandthesizeofurbanpopulationinthetotalpopulationinthesetwo placesbetween1980and2010,andthedataispresentedintable'5andtable'6. Table5:UrbanandRuralPerCapitaHouseholdIncomeandUrbanizationin Shanghai,1980(2010.(ShanghaiStatisticsBureau,2014). Year Urban per capita income (yuan) Rural per capita income (yuan) Urban/rural ratio Urban population (%)

37 Table6:UrbanandRuralPerCapitaHouseholdIncomeandUrbanizationin Yunnan,1980(2010.(YunnanStatisticsBureau,2011). Year Urban per capita income (yuan) Rural per capita income (yuan) Urban/rural ratio Urban population (%) By comparing the statistics of Shanghai with that of Yunnan, we can see that although the urban and rural household income in both provinces increased significantly between 1980 and 2010, Shanghai has been developing in a much faster speed than Yunnan since 1980s. In 1980, the difference between the urban per capita income in Shanghai and Yunnan was only 217 yuan, but by 2010, urban residents in Shanghaiearnedroughly2timesgreaterthanruralresidentsinYunnan,resultingina differenceof15,774'yuan.whatismuchworseisthegapinruralpercapitaincome,in 2010, rural residents in Shanghai earned 3.48 times greater than rural residents in 29

38 Yunnan. These observations are expected because of Shanghai s advantage in geographicallocationandthechinesegovernment spoliciesinestablishingshanghaias afinancialcenter. Also, the data on the urban population as percentage of the total population showsthatthemajorityofthepopulationinshanghailivesinurbanareas.in1980,more than 60 percent of the population in Shanghai lived in urban areas, and the number increased to 88.9 percent in In comparison, however, only 10.9 percent and 16.6 percentoftheyunnanpopulationlivedinurbanareasin1980and2010,respectively. Moreover,theurban3ruralincomegapinShanghaiissignificantlysmallerthanthatin Yunnan. Between 1980 and 1998, the urban3rural income ratio in Shanghai was lower than2,althoughtheratioreached2in1999,ithasnotgonebeyond2.3.theurban3rural incomeratioinyunnanstartedatamuchhigherlevelof2.8in1980,wentbeyond4in 1995andremainedatsuchahighlevel.In2008,theratioreached4.8,anextremelylarge gapwhencomparingwithallotherprovincesinchina.takingthesefindingstogether, evidence can be found to support the hypothesis. The relatively higher level of rural development and considerably smaller size of the rural area in Shanghai result in smallerurban3ruralincomegap.whiletheincomeofurbanresidentsinyunnanhave grown fast, rural residents, with a lower starting point in income and larger size, experienced slow progress and lagged development, enlarging the urban3rural gap. 30

39 Although the pace of rural development and the size of rural sector may explain the regional differences in the urban3rural income gap, the income gaps increased in the 1980s and 1990s regardless of the decreasing rural sector in both of the provinces, suggestingthepossibilitythatsomecommonfactorsinthisperiodwereenlargingthe urban3ruraldivideacrossprovinces. Despitetheregionaldifferencesintheurban3ruralincomegap,thedatainTable'4 alsoshowsthatthreeprovinces(sichuan,tibet,andxinjiang)experiencedcontinuous declinesintheurban3ruralincomeratios.thisisdifferentfromthetrendsatthenational levelandinotherprovinces.thedeclineinthesethreeprovincescanbeattributedtothe equalizing efforts implemented in these provinces by the central government to maintainpoliticalstabilityandpreventrevolutionarydanger.tibetandxinjiangshare twoimportantfeaturesincommon.oneisthatthemajorityofthepopulationlivesin ruralareas,theruralpopulationaccountsforover50percentand70percentofthetotal populationintibetandxinjiang,respectively(nbs,2000;2010).andtheotheristhat these two provinces are majority3minority areas, meaning that the majority of the populationinthesetwoareasbelongstoethnicminorities.accordingtothecensusdata in 2000, over 90 percent of population in Tibet and over 50 percent of population in Xinjiang were ethnic minorities (NBS, 2000; 2010). Sichuan is a province with a wide varietyofethnicminorities,althoughethnicminoritiesdonotformthemajorityofthe 31

40 populationinsichuan,thenorthwestpartofsichuanisanimportantareawhereethnic minoritiesconcentrate. The living conditions of ethnic minorities in China are consistently and significantly lower than those of the ethnic majority (or the han ethnicity). Ethnic inequalityisanotherimportanttopicinchina sinequality(forexample,seegustafsson & Shi (2003)), but this is out of the scope of this thesis. As early as in the 1950s, the Chinesegovernmentestablishedasystemofregionalautonomyinminorityareasand allowedlocalethnicminoritiestomanagetheirownregionalaffairs.tibetandxinjiang are provinces with regional autonomy. As part of the regional political system, the central government initiated numerous economic plans aiming at assisting the rural developmentandincreasingthelivingconditionsoftheethnicminoritiesinthesetwo provinces in order to prevent political conflict. All the equalizing forces in these two provinces targeting at the ethnic minorities can be the reasons behind the declining urban3ruralincomegap. 3.5OverallTrendsintheUrban(RuralGap This thesis reexamines the urban3rural gap in China by using the most recent datafrommultiplesources.almostalltheevidenceshowsthatthegapisincreasingand widening,andthereislimitedevidencethatthegapwillnarrowinthenearfuture.the sectoral income gap has been widening since China s market reform, and the urban 32

41 householdspercapitaincomeisconsistentlyover3timesgreaterthanruralincomeover thepastdecade.theeducationalandhealthcaregapshavealsobeenenlargedoverthe past few decades. The urban and rural populations differ significantly not only in income and wealth, but also in their access to education and healthcare resources. Provincialdatashowsregionaldifferenceexistintermsoftheurban3ruralincomegap, butthegeneraltrendsattheprovinciallevelovertimemirrorthoseatthenationallevel. Two case studies in Yunnan and Shanghai provide some evidence that the regional differencemaybeattributedtodifferentlevelofruraldevelopmentandthesizeofrural sector. Despite the gap presented here, the urban3rural differences also exist in other areasincludingbutnotlimitedtohousing,taxation,andpoliticalrights.bothequalizing forces from government and recent vast increase in rural3to3urban migration seem to havehadlimitedeffectsincounteractingthesegrowinggaps.therootbehindthegaps shouldbefoundinthepublicpoliciesdevelopedintheearlyyearsofdevelopmentto achieve two aims: 1) to differentiate the urban and rural populations, control the movementofpopulation,andprotectthebenefitsofurbaneliteswiththemostpolitical andeconomicresources;2)tofavorthedevelopmentofurbanareasintheearlyyearsof market reform. In the next section, I analyze the influence of these important public policiesonthewideningurban3ruralgapinchina. 33

42 4."Social"Policies"and"the"Urban(Rural&Gap TheincreasingandpersistinggapissurprisinginChina,andoneofthemajor reasons is the biased public policies developed in the early years of development in favorofadvancingeconomicgrowthinurbanareasandcontrollingpopulationmobility to achieve political stability under an authoritarian regime. The following sections attempttoanalyzethesepublicpoliciesandprovideevidenceofhowthesepoliciesare associatedwiththelargeurban3ruraldivideinchina.amongthethreesocialpolicies mentionedabove,thehouseholdregistrationsystemisthemostprominentonedueto itslonghistoryandsignificantimpactandisthemajorfocusofmyanalysis. 4.1HouseholdRegistrationSystem Thehouseholdregistrationsystemorhukouoriginatedin1951whenChinawas underastatesocialisteconomy.basiclivinggoodslikefoodandotherhouseholditems weresubjecttoarationingsystemthatdistributedrationcertificatestolocalresidentsby thegovernment.localresidentscouldthenusetherationcertificatestopurchasegoods indesignedstate3ownedstores.withtheaimofadministratingthedistributionofration certificatesandpreventingpeoplefromoneplacetopurchasegoodsinanotherplace, hukou' was first proposed. However, except controlling for the food and goods purchasingprocess,hukou'didnotlimitpeople smobilitybetweenurbanandruralareas intermsofemploymentandwasnotassociatedwithaccesstoanysocialservices. 34

43 In 1953, China developed its first Five Year Plan ( ) with the declared aim of building the foundation of socialist industrialization (Chinese Communist Party (CPC), 2015). A considerable number of state3owned enterprises were launched andavastnumberoflaborswereneeded,millionsofpreviouspeasantsfromruralareas wererecruitedwithoutrestrictions(wu&treiman,2004).however,thehugeflowof migrant workers from rural areas endangered the central government s plan of controllinghumanresourcesdistributionandbuildingawelfaresysteminurbanareas. In response, the Chinese government designed a strict hukou' system in 1955 and the NationalPeople scongress(npc),thehighestlaw3makingbodyinchina,passedthe first and only official regulation on hukou administration in 1958, which has been implementedtilltoday. Lookingattheregulation,thehukou'systemisestablishedfor maintainingsocial stability, protecting citizens rights and benefits, and serving the Chinese socialist construction (NPC, 1958). The hukou system first established two different types of registrationstatus, agricultural and non3agricultural,or urban and rural.each citizen is required to register as one of the two categories, and the initial registration statusisdeterminedbylocationofresidenceandtheoccupationoffather.inurbanareas, theregistrationunitishousehold;theruralpopulationregistersascommunesorvillages. 35

44 Oncetheinitialregistrationiscompleted,theregistrationstatusesofthenewbornsare determinedbythestatusoftheirmother. The change of hukou status is subject to strict official sanction, and one study finds that the three most important channels for rural population to convert to urban status are higher education, CPC membership, and military service (Wu & Treiman, 2004). All these channels, however, are extremely limited to the rural population, making it difficult for hukou conversion. The hukou system is the first man3made institution that created the distinction between the urban population and the rural population in China, some scholars even argue that the registration system separates Chinaintotworigidranks:urbanrankandruralrank(Renetal,1996). To support the implementation of the hukou system, the Chinese government haveadoptedcomplementarypoliciesstartingfromthelate1970s.in1980snpcpassed the Household Responsibility System (HPS) in rural areas with the declared aim to promotestructuraladjustmentinruralareas.underthissystem,state3ownedlandsin rural areas were distributed to peasants and state3planned quota on agricultural production was lifted. As a result, peasants could manage and design agricultural production by themselves. However, rural households only have the rights of use for theirlandnotrightsofalienation.oncetheypermanentlyleavetheiroriginalplaceof residence, the lands will be taken by local authorities, which results in a significantly 36

45 highcostforruralhouseholdsthatconsidermovingtourbanareasforhigherincomes andbettersocialservices(yang,1997).anotherchangeinthehukou'inthe1980swasthat local governments started to issue temporary resident permits to people with rural hukou' but who have obtained formal employment in urban areas (Liu, 2005). The permitsgrantedruralworkerstherightstoworkandliveinurbanareasbuttheydid nothaveaccesstourbaneducationresources,welfareprogramsandothersocialservices. The hukou system has important relevance to the urban3rural inequality in severalways.first,otherdevelopingcountriesalsohavehouseholdregistrationsystems, butthehukousysteminchinaisuniqueinthewaythatitdoesnotonlyaimatproviding demographicstatistics,butalsoexplicitlyatdifferentiatingtheruralpopulationfromthe urban population and controlling population movement. Second, rural3to3urban migration is strictly inhibited under the system, and other complementary policies imposeunaffordablecostsontheruralpopulationwhointendstomovetourbanareas. Althoughrural3urbanmigrationhasbeenincreasingsincetheearly1990sandmillions of previous peasants are now working in urban areas, their rights in various social servicesandaccesstourbanwelfareprogramsaredenied. Third, hukou also alienates the rural population from the major benefits of economicgrowthinurbanareas.thehukousystemnotonlyrestrictspeople slocationof residence, but also controls birth, marriage, death, employment and social welfare 37

46 including education, healthcare, and housing. For example, both state3owned and privately owned enterprises in urban areas are only allowed to recruit people with urban hukou' and welfare programs and schools in urban areas are only accessible to people with urban hukou and formal employment. Under this system, the urban and ruralpopulationsareascribeddifferentstatusandlifeopportunitiesandtheirstatusare extremelydifficulttochange. However,itismisleadingtoconcludethatthehukousystemitselfhascausedthe urban3ruraldivideinchina.thesystemshouldbeviewedasapoliticalstrategyutilized bytheurbanelitestoconcentrateresourcesontheurbandevelopment,whileexcluding the rural population from enjoying the fruits of economic development. To further provide support for these claims, I use the 1995 and 2002 Chinese Household Income Project (CHIP), a nationally representative survey on individual and household characteristicsinchina surbanandruralareasandresourcesfromotherindependent studies to show how the hukou system has influenced the income and educational inequality. Accordingtothecensusdataonhukou'registration,around25percentofChinese populationhasurbanhukoustatusin2000,andthisnumberincreasedto29percentin Hukou'is tightly correlated with place of residence. Previous studies have found thatthesimplefactofbeinginurbanareascanresultinsignificantprivilegeinincome. 38