IntergovernmentaltransferinChina.docx

IntergovernmentaltransferinChina.docx

《IntergovernmentaltransferinChina.docx》由会员分享，可在线阅读，更多相关《IntergovernmentaltransferinChina.docx（20页珍藏版）》请在冰点文库上搜索。

IntergovernmentaltransferinChina

Draft

IntergovernmentaltransferinChina

XinyeZheng1,JorgeMartinez-Vazquez2,LiZhang3

1SchoolofEconomics,RenminUniversityofChina

2InternationalStudiesProgram,AndrewYoungSchoolofPolicyStudies,GeorgiaStateUniversity

3GraduateSchoolofPublicFinanceandPublicPolicy,CentralUniversityofFinanceandEconomics,China

I.Introduction

TherehasbeenanincreasinginterestontheintergovernmentaltransfersysteminChinainrecentyears,amongotherissues.Nonetheless,comparedwithotheraspectsoftheeconomy,intergovernmentaltransfersystemisoneofthefewtopicsreceivingrelativelylessattention,partlyowingtotheunavailabilityofthedata.

LiteraturesontransfersysteminChinamainlydweltontheroleoftransfersinachievingtheobjectiveofequalizationacrosslocalities（MartinezandZhang,2002;PerssonandEriksson,2005）;andwhetherdifferentlayersofsub-nationalgovernmentstructuresenforceoroffsetthegoalofthecentralgovernments（BahlandWallace,2002;MartinezandTimofeev,2006）.Thequestionweareinterestedinis,whatfactorsdeterminehowmuchtransferseachlocalityobtains?

Inanotherword,whatcriteriadoesthecentralgovernmentorhigherlevelgovernmentfollowinallocatingtransferstolowerlevelgovernment?

ArethoseNationallyDesignatedPovertyCounties（NDPCs）receivingtransfersbasedonfactorsdifferentfromthosefornon-NDPCs?

Notuntilrecentlytheliteraturestartedtoexplorethetransfersysteminthisdirection.However,thelimitedresearchesseemtoputtheemphasisonthepoliticalfactorsinsteadofeconomicfactors.Forexample,PerssonandEriksson（2005）owethetrendoftransfersbecomingequalizedtothedesireofcentralgovernmenttoequalizetheincomedisparityinordertoachievesocialstability.Wang（2005）alsoseesthetransfersastoolsavailabletoChinesepoliticianstopleasetheirconstituentsortoneutralizepotentialthreats.

However,besidesthepoliticalconcerns,theeconomicconditionsshouldalsoplayveryimportantroles,usuallymoresignificantthanthepoliticalones.EspeciallyChinahasbeenadoptingthemarketeconomymechanism,andtherelevantreformsconductedinvarioussectors,includingthefiscalrelationshaveputmoreandmoreemphasisoneconomicaspectsinsteadofpoliticalaspects.That’swhyitiscomparativetoanalyzethedeterminantsoftransfersinChinafromtheeconomicperspective.

MostoftheliteratureontransfersysteminChinahasbeenrestrictedtotheprovincialdata,mainlyduetothedataavailability.Nowwithourdataset,weareabletoextendtheanalysistocountylevelin1997-2003.Wearehopingthatthecountyleveldatasetcouldfacilitateustocapturemoreofthevariationsintransfersacrosscounties,andexplainthedeterminantsofthetransferallocationsoractualtransferswithmoreprecision.

Ourempiricalresultsshowthat,thetotaleffectofthetransferisnotequalizing,eventhoughthetransferwithoutthetaxrebateisequalizing,theeffectsofthetaxrebateisdominating,whichispro-rich.Eventhoughchinahassetuptheobjectiveofequalization,theactualpolicyconsequenceshowstheobjectivehasnotbeenfulfilled.

Thispaperisconstructedinthefollowingway.Inthenextsection,wereviewtheeconomicsbehindtheintergovernmentaltransfer:

whyhigherlevelgovernmentsmaketransferstolowerlevelgovernmentsandhowtheyactuallyoperatethetransfersystem.InsectionIII,weintroducethetransfersysteminChina,howitevolvedoveryearandwhatareincludedinthecurrenttransfersystem.InsectionIV,weconductempiricalanalysis,usingourcounty-leveldatasettoexaminehowthetransfersareallocatedtoeachcounty.Wedothisusingthetotaltransferaswellasdifferentiatingbetweenthetwobroadcategoriesoftransfers.InsectionV,wefirstdivideoursampleintotwosubgroups,NDPCsandnon-NDPCsandseehowthingschangebetweenthesetwo.Thenwedividethewholesampleintourbancountiesandruralcounties.Finallywedrawourconclusion.

II.TheEconomicFactorsDeterminingTransfersinChineseCounties

Weconductourempiricalanalysisbasedonthesumofallcategoriesoftransfers,andalsodivideitintotaxrebateandallothertransfers,mostlyforthepurposeofequalization.Weconductfixed-effectsregressions,andtheresultsareinthefollowingtable（seeTable1）.Fromthistablewecanseethat,forallothertransfers,thehigherthepercapitaGDPlevel,thelowerthepercapitatransfersacountygets.However,thehigherthepercapitaGDP,thehigherthetaxrebates.Fortotaltransfers,therelationshipisagainpositive,whichmeanstheeffectsofthetaxrebatedominanttheeffectsoftheequalizationtransfers.Allresultsarehighlysignificant.

Similarly,wecanseethisfromanothervariable,theownrevenueinpercapitaterm:

thecoefficientsarepositiveforbothtaxrebatesandthetotaltransfer,butnegativeforequalizationtransfers;withallresultsaresignificantat1%level.Theamountsofpercapitaequalizationtransferscountiesgetarenegativelycorrelatedwithpercapitaownrevenue,whilethepercapitaownrevenueispositivelycorrelatedwiththetaxrebatesinpercapitaterm.Sumupthesetwocounteractingforces,theeffectsoftaxrebatesareagaindominant.

Onthecontrary,whenwelookatthelagofpercapitaexpenditure,thingsarereversed.Thelagofpercapitaexpenditureispositivelycorrelatedwithtransferswhetherornotwetakeintoconsiderationofthetaxrebates:

thehighertheexpenditureinlastperiod,significantlythehigherthetransfersareallocatedtothecountythisperiod.However,whenitcomestotaxrebatesinpercapitaterm,therelationshipbecomesnegativeandsignificant.Sincethetaxrebatesarecalculatedbasedontheformulawedenotedabove,thisnegativerelationshipshouldn’tbeinterpretedasthelowertheexpenditurelastyear,thehighertherebatethisyear.

Inaddition,thepopulationvariablehasnegativesignexceptforrebates,butnoneofthemissignificant.However,theshareofruralpopulationintotalispositiveandsignificantforequalizationtransfers,meaningthatthehighertheproportionofruralpopulation,themoreequalizationtransfersareallocated.Thisconfirmsthatequalizationtransfersarepro-poor.Buttheeffectsarenotpresentinthecaseoftaxrebates;inregression（4）thesignevenbecomesnegative.Fortunately,thistimethepro-poorfeaturefromtheequalizationtransferisdominantsincefortotaltransfers,theruralpopulationshareisagainpositiveandsignificant.

Fromthetimedummies,wecanseetheamountsoftheequalizationtransferandtotaltransferarebothincreasingoveryears,whilethetaxrebatedidnotexhibitincreasingtrenduntil2002.

Table1Transfersresultsforthewholesample

（1）

（2）

（3）

（4）

（5）

（6）

AllinPerCapitaTerm

EqualizationTransfer

EqualizationTransfer

TaxRebate

TaxRebate

TotalTransfer

TotalTransfer

PerCapitaGDP

-0.001

-0.001

0.007

0.006

0.006

0.006

（2.21）**

（2.75）***

（46.61）***

（44.54）***

（19.54）***

（17.90）***

Population

-0.221

-0.220

0.045

0.115

-0.176

-0.106

（1.54）

（1.57）

（0.63）

（1.64）

（1.13）

（0.69）

LagofPercapitaExp

0.241

0.244

-0.019

-0.011

0.221

0.233

（39.91）***

（40.15）***

（6.27）***

（3.70）***

（33.73）***

（35.37）***

ShareofRuralPopulation

1.323

1.214

0.090

-0.129

1.412

1.086

（4.68）***

（4.29）***

（0.63）

（0.91）

（4.59）***

（3.54）***

ShareofPublicEmployee/Pop

591.989

1,428.969

2,020.958

（3.89）***

（18.90）***

（12.27）***

PerCapitaOwnRevenue

-0.035

-0.039

0.172

0.164

0.137

0.126

（5.28）***

（5.75）***

（51.16）***

（49.14）***

（18.83）***

（17.24）***

y98

-231.754

-232.098

-19.486

-20.802

-251.241

-252.900

（44.54）***

（44.65）***

（7.43）***

（8.05）***

（44.32）***

（44.92）***

y99

-207.597

-205.218

-24.435

-20.173

-232.032

-225.391

（41.64）***

（41.24）***

（9.72）***

（8.16）***

（42.72）***

（41.82）***

y00

-160.795

-159.270

-27.493

-24.621

-188.288

-183.891

（33.65）***

（33.34）***

（11.41）***

（10.37）***

（36.17）***

（35.54）***

y01

-78.004

-76.811

-36.732

-34.464

-114.737

-111.275

（16.96）***

（16.70）***

（15.84）***

（15.08）***

（22.90）***

（22.34）***

y02

-29.098

-28.795

4.315

4.794

-24.783

-24.001

（6.72）***

（6.66）***

（1.98）**

（2.23）**

（5.26）***

（5.12）***

FiscalDependentsTotal

-0.000

0.001

0.000

（0.65）

（2.03）**

（0.35）

Constant

174.474

159.247

3.802

-22.236

178.276

137.010

（7.32）***

（6.72）***

（0.32）

（1.89）*

（6.86）***

（5.34）***

Observations

14433

14433

14433

14433

14433

14433

NumberofID

2703

2703

2703

2703

2703

2703

R-squared

0.45

0.45

0.46

0.48

0.52

0.53

Therefore,fromtheresultsofourregressions,thecentralgovernmentdoestrytoengageintheequalizationefforts,asexemplifiedbytheequalizationtransfers.However,duetotheexistenceofthetaxrebates,thisequalizationefforthasbeenmorethanoffset.Theoveralleffectofthetransfersystembecomespro-rich.Theobjectiveofthepolicyhasnotbeenfulfilled.

III.Extension

（1）TheTransferstoNDPCsandnon-NDPCs

Inthissection,wedividethewholesampleintotwosub-groups:

theNationallyDesignatedPovertyCounties（NDPCs）andthosenon-NDPCs.WeruntheregressionsseparatelyforthesetwogroupsandtheresultsareinthefollowingTable2.

Fromthistablewecanseethat,forNDPCs,thingsaresignificantlydifferentfromthecasewiththewholesample.First,thepercapitaGDPdoesn’texhibitanysystematicrelationshiptothetwopartsoftransfersaswellasthetotaltransfer.Thesignsofthecoefficientsoftheequalizationtransfersandtaxreb