Foreign direct investment in China_a spatial econo

WORKINGPAPERSERIESForeignDirectInvestmentinChina:ASpatialEconometricStudyCletusC.CoughlinEranSegevWorkingPaper1999-001A(otherthananacknowledgmentthatthewriterhashadaccesstounpublishedmaterial)shouldbeclearedwiththeauthororauthors.PhotocourtesyofTheGatewayArch,St.Louis,MO.:ForeignInvestment,EDTLocation,China,SpatialEconometrics,Foreign-OwnedFirmsJELCLASSIFICATIONS:053,F21,R12,R30CletusC.CoughlinEranSegevVicePresidentSeniorResearchAssociateFederalReserveBankofSt.LouisFederalReserveBankofSt.Louis411LocustStreet411LocustStreetSt.Louis,MO63102St.Louis,MO63102ForeignDirectInvestmentinChina:ASpatialEconometricStudy(RunningTitle:ForeignDirectInvestmentinChina)CletusC.Cough/inandEranSegev1.INTRODUCTIONAftersealingitselffordecadesfromtheglobaleconomy,inthelate1970sChinabegantoremovesomeofthebarrierstotheinflowofforeigndirectinvestment.However,thelackofprecedent,coupledwithanuncertainpoliticalclimateandotherunfavorablefactors,atfirstseverelyhinderedChineseattemptstoattractFDI.In1980,theflowofFDIintoChinatotaledlessthan$200million(USdollars).Tn1997,however,theflowofFDIexceeded$44.9billion,morethan225timeslargerthantheflowin1980.ThatfiguremadeChinathelargestrecipientofFDIamongdevelopingcountries,andthesecondlargestintheworld(aftertheUnitedStates).CLETUSC.COUGHLINi~vicePresidentandAssociateDirectorofResearchattheFederalReserveBankofSt.Louis.BRANSEGEV,formerlySeniorResearchAssociateattheFederalReserveBankofSt.Louis,isagraduatestudentattheJohnF.KennedySchoolofGovernmentatHarvardUniversity.TheauthorsaregratefultoJ.RayBowen,ChenChunlai,GilbertoEspinoza,JeffCohenandtwoanonymousrefereesfortheirhelpfulcommentsandsuggestions.1TheseflowsofFDIareplaying,andwilllikelycontinuetoplay,akeyroleintheintegrationofChinaintotheworldeconomy.ThefutureofChinesestate-ownedenterprises,aswellasthecountry’seconomicdevelopmentgenerally,iscloselyrelatedtoFDIactivity.Numerouspoliticalandeconomicissueswillariseindeterminingthefateofinefficientstate-ownedenterprises.Henleyetal.(1999)arguethatlocalgovernmentsinChinawillbekeyplayersinresolvingtheseissues.1OurresearchfocusesonthegeographicdistributionofFDIwithinChina,thebulkofwhichhasbeendirectedtoregionsalongthecoast.FactorsaffectingthelocationofFDIcanprovideguidancetopolicymakersinidentifyingtheobstaclesthatsomeregionsmustovercometoattractFDI.Usingprovincialdata,weconstructandestimateamodeltoexplainthegeographicpatternofFDIlocationwithinChinasince1990,wheninflowsofFDIbegantoincreaserapidly.OurresearchbuildsuponthatofBroadmanandSun(1997),ChunlaiChen(1997d)andChien-HsunChen(1996).InadditiontousingmorerecentFDIdataandtestingtheexplanatorypowerofadditionalvariables,weextendpreviousresearchinoneespeciallynoteworthydimension.Previousresearchutilizesstandardeconometricmodelsand,thus,failstoincorporatetheWhilelocalgovernmentsinChinasofarappeartobecompetingtoattractFDI,BranstetterandFeenstra(1999)suggestthatChinesestate-ownedenterprisesarelikelytoviewforeign-ownedenterprisesascompetitivethreatsand,thus,opposetheentryofforeignfirms.2distinctivecharacteristicsofspatialdata.Weincorporatethespatialcharacteristicsofthedatainouranalysis.Withspatialdatathelocationofobservationsisanimportantattribute,becauseneighboringregionscanaffectoneanother.Manyreasonscanbeprovidedforwhatistermedspatialdependence.InthecaseofFDI,agglomerationmayleadtohigherFDIlevelsinneighboringprovincestotheextentthatitsbeneficialeffectsspilloverprovinceborde