1、原文DETERMINANTSOFFOREIGNDIRECTINVESTMENTATTHEREGIONALLEVELINCHINAASTHEWORLDSMOSTPOPULATEDCOUNTRY,CHINAHASATTRACTEDAGREATDEALOFATTENTIONFROMAWIDERANGECOMPANIESSEEKINGTOLEVERAGETHELOWRELATIVECOSTOFEMPLOYINGACHINESEWORKFORCE,ORTOGAINACCESSTOTHEGROWINGCHINESEMIDDLEANDUPPERCLASSESFROMTHEINVESTORSPOINTOFVI
2、EW,PICKINGTHERIGHTLOCATIONTHATPROVIDESACOMPETITIVEADVANTAGEISCRITICALTODEVELOPINGASUSTAINABLEBUSINESSMODELTHELOCATIONHEORSHEWOULDCHOOSEASTHEDESTINATIONOFFDIMUSTULTIMATELYBEMOREPROFITABLETOINVESTINTHANINOTHERSCOUGHLINETAL,1991CHOOSINGTHERIGHTLOCATIONOFTENINVOLVESMANYDIFFERENTFACTORSHYMER1960FOUNDTHAT
3、AMERICANFDIWASMAINLYCONCENTRATEDINAFEWINDUSTRIESANDMONOPOLIZEDBYSEVERALCOMPANIESMULTINATIONALCOMPANIESMNCSWERETHEPRODUCTOFIMPERFECTMARKETSANDMONOPOLYADVANTAGESWHERETHECOMPANIESHADTHEADVANTAGEWITHREGARDSTOCHOOSINGWHERETOINVESTANUMBEROFCONCLUSIONSCANBEDRAWNFROMHYMERSANALYSISTHATHELPSFRAMEUPTHISSTUDYFI
4、RST,FDITENDSTOFLOWINTODIFFERENTIATEDMARKETSWHEREAMNCBELIEVESTHEYWILLHAVEANADVANTAGECOMPETITIVELYSECOND,COMPANIESTHATAREABLETOMAKEINVESTMENTSOVERSEASALLHAVECERTAINADVANTAGES,SUCHASECONOMIESOFSCALE,DIFFERENTIATEDPRODUCTS,SPECIALSKILLS,ANDLOWCOSTPRODUCTIONTHESECOMPANIESWILLMAKEINVESTMENTSINREGIONSTHATD
5、ONOTHAVETHESEADVANTAGESTHIRD,THEREAREMANYWAYSINWHICHMNCSCANINVESTOVERSEASINSUCHASEXPORTING,ANDLICENSING,INADDITIONTODIRECTINVESTMENTMNCSWITHOUTLOCALPARTNERSALWAYSPREFERTOCHOOSEFOREIGNDIRECTINVESTMENTLASTBUTNOTTHELEASTHYMERFOUNDTHATABOUTHALFOFTHEOVERSEASOPERATINGCAPITALOFAMERICANFIRMSCAMEFROMHOSTCOUN
6、TRIESTHUSFDITENDSTOFLOWINTOTHECOUNTRIESORREGIONSTHATHAVEDEVELOPEDFINANCIALSYSTEMSANDCAPITALMARKETSTHECASEFORCHINAASAHOSTFORFDIHASINCREASEDSUBSTANTIALLYSINCEHYMERFIRSTPUBLISHEDHISTHOUGHTSIN1960CHINAASASOURCEOFLOWCOSTLABOR,ANDWITHANATTRACTIVE,RAPIDLYGROWINGDOMESTICMARKETREPRESENTSGROWTHFORMANYCOMPANIE
7、STHEFACTTHATTHECOASTALREGIONSESPECIALLYTHOSEINCLOSEPROXIMITYTOHONGKONG,ANDWITHHISTORICTIESTOTHEWESTANDTAIWANALSOHAVESOPHISTICATEDFINANCIALSYSTEMSANDCAPITALMARKETSSUPPORTSHYMERSCONCLUSIONSFIGURE1OTHERFACTORSMAYALSOCONTRIBUTETOTHEEMERGENCEOFCHINAASALEADINGHOSTFORFDIINTHE1970SANDEARLY1980S,BUCKLEYANDCA
8、SSON1976ANDRUGMAN1981PUTFORWARDTHECONCEPTOFINTERNALIZATIONINTERNALIZATION2MEANSTHEPROCESSOFESTABLISHINGAMARKETINSIDETHECOMPANYANDSUBSTITUTINGTHEINTERNALMARKETFORANEXTERNALMARKETTHETRANSFERPRICEINSIDETHECOMPANYENABLESTHEINTERNALMARKETTOOPERATEASEFFECTIVELYASTHEEXTERNALMARKETTHETHEORYHOLDSTHEOPINIONTH
9、ATTHEIMPERFECTIONOFEXTERNALMARKETSCOMPELSTHECOMPANYTOEXCHANGECERTAINPRODUCTSINSIDETHECOMPANYWHENTHISHAPPENSACROSSNATIONALBOUNDARIES,ITACTSASFDIOFMNCSCOUNTRIESTHATBRINGABOUTTHEIMPERFECTIONSINTHEEXTERNALMARKETSSUCHASHIGHTARIFFOFNONTARIFFBARRIERSAREWHEREFDITENDSTOFLOWTHETRENDTOWARDCOLLABORATIONTHROUGHJ
10、OINTVENTURES,PARTNERSHIPS,ANDOTHERFORMSOFCOLLABORATIVERELATIONSHIPSMAYSUGGESTTHATCOMPANIESAREATTEMPTINGTOCREATE“INTERNALMARKETS”WITHASELECTGROUPOFCOMPANIESTHECENTRALANDREGIONALGOVERNMENTSTHROUGHTHEDEGREEOFOPENNESSIEFOREIGN,PRIVATEOWNERSHIPOFSTATEOWNEDENTERPRISESMAYBESIGNIFICANTLYRESPONSIBLEFORTHECRE
11、ATIONOFTHESE“INTERNALMARKETS”WHILELOWCOST,ACCESSTOTHEDOMESTICMARKETS,ANDTHECREATIONOFINTERNALMARKETSTHATSIMPLIFYTHETRANSFEROFFDIAREOFTENCOMPELLINGFROMACOUNTRYPOINTOFVIEWFROMAREGIONALPERSPECTIVE,SELECTINGTHERIGHTLOCATIONCANBEASIGNIFICANTCHALLENGEDUNNING1977ASSUMEDTHATTOMAKEAFDI,COMPANIESSHOULDHAVETHR
12、EEADVANTAGESANOWNERSHIPADVANTAGEALOCATIONADVANTAGEANDANINTERNALIZATIONADVANTAGEINFLUENCEDBYINDUSTRYLOCATIONTHEORY,DUNNINGDETERMINEDTHEDEFINITIONOFLOCATIONADVANTAGEASTHEADVANTAGEONELOCATIONHADOVERANOTHERBASEDUPONNATURALANDHUMANRESOURCESTHEPRICE,QUALITYANDPRODUCTIVITYOFINPUTSINTERNATIONALTRANSPORTATIO
13、NANDCOMMUNICATIONCOSTINVESTMENTFAVORORDISCRIMINATIONMANMADEBARRIERSTOTRADEFUNDAMENTALFACILITIESCROSSNATIONALVALUES,LANGUAGE,CULTURE,COMMERCIALPRACTICEANDPOLITICSRESEARCHANDDEVELOPMENTTHECONCENTRATIONOFPRODUCTIONANDSALESECONOMICSYSTEMANDPOLITICALSTRATEGYRESOURCEALLOCATIONSYSTEMPP395418THELOCATIONADVA
14、NTAGEOREDGETHEORYISRELEVANTTOTHISSTUDYASITHOLDSTHEOPINIONTHATTHEREASONWHYANINTERNATIONALCOMPANYMAKESDIRECTINVESTMENTSINAPARTICULARREGIONISTHATTHECOMPANYWOULDLIKETOATTAINCERTAINLOCATIONADVANTAGESWHICHDONOTEXISTINOTHERHOSTCOUNTRIESVARIOUSSTUDIESTABLEIHAVEHIGHLIGHTEDTHEFACTORSTHATINVESTORSCAREABOUTINCL
15、UDINGLABORCOST,MARKETSIZEANDMARKETPOTENTIAL,TRADEBARRIERSANDCOUNTRYRISKSTHOSEHOSTCOUNTRIESWITHLOWERLABORCOST,LOWERTRANSPORTCOST,GREATERMARKETPOTENTIAL,TRADEPROTECTIONISM,SMALLERCOUNTRYRISK,BETTERINFRASTRUCTUREANDBETTEREDUCATEDANDSKILLEDPOPULATIONSARETYPICALLYTHEFOCUSOFFDIACCORDINGTOTRADITIONALLOCATI
16、ONTHEORY,INVESTMENTINCENTIVESCANBEMAINLYDIVIDEDINTOTWOTYPESFIRST,THOSETARGETEDATCOSTSAVINGS,WHICHPURSUEAPRODUCTIONCOSTEDGEINHOSTCOUNTRIESWHEREFDIISMAINLYFOCUSEDONPRODUCINGLOCALLYANDTHENEXPORTINGANDSECOND,FORTHOSECOMPANIESWHOAIMTOEXPANDTHEIRMARKETPRESENCETHROUGHINCREASINGTHEIRPENETRATIONIN“LOCAL”MARK
17、ETSFDIMAINLYFOCUSESONLOCALPRODUCTIONANDLOCALSALEASOPPOSEDTOEXPORTING,SOTHISKINDOFFDIPLACESAHIGHEMPHASISONMARKETSIZE,MARKETGROWTH,ANDCONSUMPTIONABILITYEMPIRICALWORKSLOCALMARKETSAREOFTENCONSIDEREDBYCOUNTRYORBYREGIONTHEREISAVASTAMOUNTOFEXISTINGLITERATUREFOCUSEDONTHEGEOGRAPHICALDISTRIBUTIONOFFDITABLEIIN
18、UMEROUSSTUDIESHAVEBEENDONETHATIDENTIFYTHEDETERMINANTSOFFDIFORREGIONS,COUNTRIES,REGIONSWITHINCOUNTRIES,ASWELLASFDIFLOWSBETWEENCOUNTRIESWHILETHEREISNOTNECESSARILYASINGLESETOFFACTORSTHATCONSISTENTLYDETERMINEFDI,INFRASTRUCTURE,SKILLEDLABOR,LABORCOST/WAGE,AGGLOMERATION,ANDGOVERNMENTSUPPORTINTERMSOFOPENNE
19、SS,AND/ORATTEMPTSATATTRACTINGFDIAREOFTENCITEDASHAVINGASIGNIFICANTRELATIONSHIPBETWEENFDIANDTHELOCATIONCHOICEANDWHILEREGIONSTHATRECEIVESUBSTANTIALFDIMAYBENEFIT,ITISULTIMATELYTHELINKAGESBETWEENEDUCATION,GOVERNMENT,ANDINDUSTRYTHATPERPETUATETHEDEVELOPMENTOFNEWACTIVITIESINTHEREGION,LEADINGTOINNOVATION,AND
20、FURTHERINVESTMENTPORTERANDSTERN,2001WITHTHEFASTDEVELOPMENTOFITSECONOMY,CHINAHASBEENOFCRUCIALIMPORTANCETOFDI,THUSTHESTUDYOFINVESTMENTSINCHINAHASBEENGROWINGTABLEIIIWANGANDSWAIN1995EXAMINETHEDETERMINANTSOFFDIINCHINA,FINDINGTHATFDIINTHEMANUFACTURINGSECTORISPOSITIVELYRELATEDTOCHINASGDP,GDPGROWTH,WAGES,AN
21、DTRADEBARRIERS,BUTNEGATIVELYRELATEDTOINTERESTRATEANDEXCHANGERATEBHAGWATIANDSRINIVASAN1983CONDUCTEDASURVEYTORANKPROVINCESOFCHINAWITHTHEBESTINVESTMENTENVIRONMENTTHEYIDENTIFIEDANUMBEROFVARIABLESMARKETSIZEWAGE,EDUCATION,EXTENTOFINDUSTRIALIZATION,INFRASTRUCTURETRANSPORTFACILITIES,COMMUNICATIONFACILITIES,
22、LIVINGENVIRONMENT,ANDTHELEVELOFSCIENTIFICRESEARCHCHEN1996DIVIDEDCHINAINTOEAST,CENTRALANDWESTREGIONS,ANDFOUNDTHATWAGEHASNORELATIONWITHSPATIALDISTRIBUTIONOFFDIBUT,THATMARKETSIZEHASTHEMOSTINFLUENCEONTHECENTRALAREAWEI1999FOUNDTHATBRIBERYANDOTHERDISHONESTBEHAVIORSOFGOVERNORSINFLUENCEDTHEINFLOWOFFDILUMING
23、HONG2000IDENTIFIEDSOMEADDITIONALDETERMINANTSBESIDESGENERALECONOMICFACTORSSUCHASANINFRASTRUCTUREFACTORWITHENERGYCONSUMPTIONASPROXYLABORQUALITYWITHTHELITERACYPERCENTAGEASPROXYANDASYSTEMFACTORWITHHONESTGOVERNMENTANDDEGREEOFOPENNESSASPROXYBHAGWATIANDSRINIVASAN1983,ANDSUNETAL2002FOUNDACORRELATIONBETWEENI
24、NVESTMENTINSCIENTIFICRESEARCHANDFDIFLOWSTONGANDYUETING2000CONSIDEREDTHEHOMETOWNCONNECTIONSOFOVERSEASCHINESEINVESTORSASADETERMINANTOFFDIFLOWINTOCHINA,NOTINGTHATTAIWANESEINVESTORSHAVEOFTENINVESTEDTHROUGHHONGKONGTOLIMITDELAYSDUETOGOVERNMENTANDBUREAUCRACYSUNETAL2002,P88IDENTIFIEDPOSSIBLEDETERMINANTSOFFD
25、IFLOWWITHINCHINAINCLUDINGMARKETDEMANDANDSIZE,AGGLOMERATION,LABORQUALITY,LABORCOST,LEVELOFSCIENTIFICRESEARCH,DEGREEOFOPENNESS,COUNTRYRISKEMPIRICALFRAMEWORKTHELOCATIONOFFDIINCHINAHASSTRIKINGSPATIALCHARACTERISTICSTHEPROVINCESINCHINAARECLASSIFIEDINTOTHREEREGIONSEASTERN,CENTRAL,ANDWESTERNFDIISUNEVENLYDIS
26、TRIBUTEDACROSSTHETHREETABLEIVSHOWSTHATIN2002,THEEASTERNREGIONRECEIVEDMORETHAN86PERCENTOFTHETOTALFDIAMOUNTWHILETHECENTRALANDWESTERNREGIONSTOGETHERRECEIVEDLESSTHAN14PERCENTTHISISCONSISTENTWITHPASTTRENDSFORFDIINCHINAANDHASBEENCITEDASONEOFTHEREASONSTHATHAVELEDTOTHEFASTDEVELOPMENTOFTHECOASTALPROVINCESAND
27、THEWIDENINGGAPINTERMSOFECONOMICDEVELOPMENTBETWEENTHECOASTALANDINLANDPROVINCESCHENGANDZHANG,1998MOREOVER,THEFDIFLOWINTOEASTERNCHINAHASCONTINUEDTOINCREASEFROM2001,THEAMOUNTOFFDIINTHEEASTERNREGIONSHASINCREASEDBY119PERCENTWHILEINTHECENTRALREGIONS,FDIHASONLYRISENBY02PERCENTINTHEWESTERNREGIONS,FDIACTUALLY
28、DECREASEDBY139PERCENTLU,2003DFROMTHESEFIGURESITISCLEARTHATTHELOCATIONOFFDIINCHINAISCHARACTERIZEDBYENORMOUSSPATIALDIVERSITYTHEREAREOTHERTWOMAINCHARACTERISTICSOFCHINASFDIPATTERN,WHICHMAYHAVESOMECERTAINLINKSWITHTHESPATIALPATTERNTABLEIVSHOWSTHATINVESTMENTACTIVITYTENDSTOCONCENTRATEONSECONDARYINDUSTRIESLI
29、KEUTILITIES,MANUFACTURING,ANDPROPERTYDEVELOPMENT,WHILETHEPRIMARYSECTORATTRACTSONLYABOUTONEFOURTHOFTHETOTALFDITABLEIVALSOSHOWSTHATFOREIGNCAPITALFLOWSMAINLYFROMASIANCOUNTRIES,WITHMORETHAN60PERCENTOFTHETOTALFOREIGNCAPITALCOMESFROMHONGKONG,TAIWAN,JAPAN,KOREANANDOTHERSOUTHEASTASIANCOUNTRIESINTERMSOFTHETO
30、TALACCUMULATIVEAMOUNTOFFOREIGNCAPITALTHATTHECOUNTRYORAREAACTUALLYUSED100MILLION,THETOPFOURSOURCESOFFDIWEREHONGKONG2,04875MN,THEUSA39889MN,JAPAN36340MN,ANDTAIWANPROVINCE33110MNLU,2003DTHESEFOURSOURCESOFFDICONTRIBUTEMORETHANTWOTHIRDSOFTOTALFDIHOWEVER,HONGKONGSSHAREOFFDIFAREXCEEDSTHOSEOFOTHERCOUNTRIESA
31、NDREGIONSFIGURE2PARTOFTHISMAYBEEXPLAINEDTOTHEPASSTHROUGHINVESTMENTSMADEBYINVESTORSFROMTAIWANANDOTHERCOUNTRIESTOEASETHEFLOWOFCAPITALATTHEREGIONALLEVEL,THEDATASHOWSTHATDISTRIBUTIONOFFDIACROSSCHINASREGIONSISVERYUNEVENIN2002,THETOPFOURAREASAREGUANGDONG,JIANGSU,SHANDONG,ANDSHANGHAI,WHICHACCOUNTEDFOR5928P
32、ERCENTOFTHETOTALFDIFIGURE3LU,2003A,B,CTHEREASONSFORFDIBYREGIONAREASSTATEDPREVIOUSLYVARIEDINTHEEASTERNREGIONS,ITHASBEENARGUEDTHATPROXIMITY,HISTORY,CULTURE,ANDLANGUAGEALLPLAYASIGNIFICANTROLEINDETERMININGFDIOTHERSTABLEIIIHAVEFOUNDAWIDERANGEOFOTHERSIGNIFICANTDETERMINANTSOFFDIFROMTHESEPASTANALYSESOFPOTEN
33、TIALDETERMINANTSOFFDIINCHINA,FIVEWERECHOSENTODETERMINEIFTHEREWASANYSIGNIFICANTRELATIONSHIPBETWEENTHEMANDFDIACROSSTHEREGIONSWITHINCHINATABLEVTABLEVIREVEALSTHEDISTRIBUTIONSOFFDIACCORDINGTOTHESEDETERMINANTSTABLEVIITHEFIRSTDETERMINANTISTHEMARKETSIZEGDPITDIRECTLYAFFECTSTHEEXPECTEDREVENUEOFTHEINVESTMENTIN
34、FACT,ONEMAJORMOTIVATIONFORFDIISTOLOOKFORNEWMARKETSSHAPIRO,1998THELARGERTHEMARKETSIZEOFAPARTICULARREGION,THEMOREFDITHEREGIONSHOULDATTRACTGIVENOTHERTHINGSREMAINCONSTANTKRAVISANDLIPESEY1982,BLOMSTROMANDLIPSEY1991ANDOTHERSHAVEIDENTIFIEDMARKETSIZEASHAVINGAPOSITIVEIMPACTONFDIWEUSEGDPPERCAPITAFORDEMANDANDS
35、IZEEFFECTTHESECONDDETERMINANTISAGGLOMERATIONROADWHICHREFERSTOTHECONCENTRATIONOFECONOMICACTIVITIESTHATLEADSTOPOSITIVEEXTERNALITIESANDTHEECONOMIESOFSCALECOUGHLINETAL1991,WHEELERANDMODY1992ANDBRAUNERHJELMANDSVENSSON1996AMONGSTOTHERSFOUNDTHATTHELEVELOFAGGLOMERATIONWASPOSITIVELYRELATEDTOTHEFDIINAPARTICUL
36、ARCOUNTRYTHISSTUDYUSESINFRASTRUCTUREQUALITYTOCAPTURETHEAGGLOMERATIONBENEFITSTHEHIGHWAYANDRAILWAYMILEAGEPERSQUAREKILOMETERISPROXYFORTHEQUALITYOFINFRASTRUCTURETHETHIRDDETERMINANTISLABORQUALITYSCHNTHISSTUDYUSEDTHETOTALNUMBEROFTHEPRIMARY,ANDSECONDARYSCHOOLS,ASWELLASUNIVERSITIESASAPROXYFOREDUCATIONANDFUR
37、THERFORLABORQUALITYMODYANDSRINIVASAN1998,LUMINGHONG2000ANDAKINLO2004,ALLFOUNDTHATLABORQUALITYHASAPOSITIVEIMPACTONFDITHEFOURTHDETERMINANTISLABORCOSTWAGE,ASMEASUREDBYWAGEBHAGWATIANDSRINIVASAN1983,COUGHLINETAL1991,WANGANDSWAIN1995ALLFOUNDARELATIONSHIPBETWEENWAGEORLABORCOST,ANDFDIHOWEVER,LABORCOSTMAYHAV
38、EANEGATIVECORRELATIONMULTINATIONALFIRMSINCHINATENDTOHIREQUALITYWORKERSWHOEARNHIGHERWAGESASAPOSSIBLEREFLECTIONOFTHISHIGHERLABORQUALITYHENCE,WAGESINTHOSEPROVINCESTHATATTRACTMOREFDIMAYBEHIGHERTHEFIFTHDETERMINANTISTHEDEGREEOFOPENNESSANDPROGRESSOFREFORMSHARESUNETAL2002ANDFUJITAANDHU2001FOUNDTHATASIGNIFIC
39、ANTRELATIONSHIPBETWEENTHEDEGREEOFOPENNESSASDEFINEDBYTHEPERCENTAGEOFSTATESOWNEDENTERPRISESSOESANDFDIONTHEONEHAND,AMOREOPENECONOMYMEANSTHATFOREIGNINVESTORSAREMOREFAMILIARWITHTHEHOSTECONOMYANDMAYTHEREFOREBEMOREWILLINGTOINVESTINTHECOUNTRYONTHEOTHERHAND,OPENNESSCANHAVEANEGATIVEIMPACTONFDIASITMAYATTRACTMO
40、RECOMPETITION,LESSENINGANYCOMPETITIVEADVANTAGEAFIRMMAYHAVEHOPEDTOREALIZEWEUSETHESHAREOFSTATEOWNEDENTERPRISESINALLENTERPRISESOFAREGIONTOMEASUREITSDEGREEOFOPENNESSANDPROGRESSOFREFORMASNOTEDPREVIOUSLY,THEREAREMANYOTHERCOMMONLYUSEDDETERMINANTSTHATMAYHAVEACORRELATIONWITHFDIINCHINADETERMINANTSLIKETHENUMBE
41、ROFTELEPHONESETS,NUMBEROFTOURISTS,LEVELOFSCIENTIFICRESEARCH,DEGREEOFINDUSTRIALIZATION,AGGLOMERATIONOFFDI,PROMOTIONEXPENDITURESFORATTRACTINGFDI,TAXSTRUCTURE,ANDTHESPECIALTREATMENTOFFEREDTOFOREIGNINVESTORSTHATMAYHAVEIMPACTS,TOOHOWEVER,SUCHDATAISMOREDIFFICULTTOOBTAINONAREGIONALORPROVINCIALLEVELINCHINA,
42、WHICHLEDTOTHEFOCUSONTHEFIVEAFOREMENTIONEDDETERMINANTSTHETYPICALMETHODOFESTIMATINGTHEEFFECTOFPOTENTIALDETERMINANTSOFFDIISTOREGRESSTHECHOSENDEPENDENTVARIABLE,SUCHASTHEAMOUNTOFFDIINALOCATION,ONASETOFINDEPENDENTVARIABLESWHICHWOULDPOSSIBLYAFFECTTHEPROFITABILITYOFINVESTMENTTHESEVARIABLESTYPICALLYREFLECTLO
43、CALMARKETPOTENTIAL,ANDTHECOSTOFPRODUCTIONANDTRANSPORT,ASWELLASTHEGENERALENVIRONMENTFACEDBYTHEMULTINATIONALCOMPANYONTHEBASISOFTHEEXISTINGSTATISTICALANALYSESOFTHELOCATIONOFFDIINCHINA,WEPOSTULATETHATFORTHEAMOUNTOFFDIINREGIONIEQUATION1WHEREGDP,ROAD,SCHN,WAGE,ANDSHAREREPRESENTTHEVARIABLESFORMARKETSIZEAND
44、DEMAND,AGGLOMERATION/INFRASTRUCTURE,LABORQUALITY,WAGE,DEGREEOFOPENNESSANDPROGRESSOFREFORM,RESPECTIVELYSINCE,OURDEPENDENTVARIABLEISTHEPERCAPITAAMOUNTOFFDI,WEUSEPERCAPITAGDPTOCAPTURETHEREGIONALMARKETPOTENTIALORMARKETSIZEINORDERTOEXPERIMENTWITHANAPPROPRIATECHOICEOFTHEINFRASTRUCTUREVARIABLEROAD,THREEALT
45、ERNATIVEPROXIESWERETESTEDTHETOTALLENGTHSOFROADSPERUNITOFLANDMASSTHETOTALLENGTHSOFHIGHGRADEPAVEDROADSPERUNITOFLANDMASSANDTHETOTALLENGTHSOFRAILWAYPERUNITOFLANDMASSAREGIONSLABORCOSTWAGEISGIVENBYITSAVERAGELABORCOSTDIVIDEDBYITSRETAILPRICEINDEXOUREDUCATIONVARIABLESCHNISTHETOTALNUMBEROFPRIMARY,SECONDARYSCH
46、OOLS,ANDUNIVERSITIESFORAGIVENREGIONALLREALVARIABLESAREMEASUREDINCURRENTPRICESINOURSAMPLE,AREGIONISEITHERAPROVINCE,CENTRALLYADMINISTEREDMUNICIPALITY,ORANAUTONOMOUSREGIONASTHEDETERMINANTSOFFDIDISTRIBUTIONWEREEXAMINEDACROSSREGIONSREGARDLESSOFTIMETRENDSANDWITHKNOWNDATALIMITATIONS,THISASSIGNMENTONLYREQUI
47、RESORDINARYLEASTSQUARESOLSESTIMATIONTHEDATAUSEDINTHISSTUDYAREOBTAINEDFROMCHINASSTATISTICALYEARBOOK2003,THECHINASTATISTICALYEARBOOKFORREGIONALECONOMY2003,THECHINAFOREIGNECONOMICSTATISTICALYEARBOOK2003,ANDTHEWHITEBOOKSOFCHINASTRADEANDECONOMICS2003FORTHISSTUDY,TWOELEMENTSOFFDIWEREUSED“SIGNEDAGREEMENT”A
48、ND“ACTUALLYUTILIZED”THELATERONEISTHEACTUALAMOUNTOFFDIINVESTEDINTHEREGIONSINCE,GDPANDWAGEAREDENOMINATEDINRMBANDTHEFDIINUSDOLLARS,WECONVERTEDTHEFDIINTORMBUSINGTHEAVERAGEYEARLYEXCHANGERATEIN2002ITISPOSSIBLETHATAHIGHCORRELATIONBETWEENTHEVARIOUSPROXIESMAYCOMEOUTANDTHATTHEPROXIESMAYOVERLAPWITHONEANOTHER,W
49、HICHMAYLEADTOSERIOUSMULTICOLLINEARITYINORDERTOCOPEWITHMULTICOLLINEARITY,THOSEHIGHLYCORRELATEDPAIRSWEREEXCLUDEDSPECIFICALLY,ALLVARIABLESWERETRANSFORMEDINTOTHENATURALLOGARITHMFORMANDWERETHENSTACKEDUPACROSSTHE30REGIONSINSUMMARY,FOREIGNDIRECTINVESTMENTATTHEREGIONALLEVELINCHINAWASEXPECTEDTOBEAFFECTEDBYTHEREGIONSMARKETDEMANDANDMARKETSIZEGDP,INFRASTRUCTUREROAD,LABORQUALITYSCHN,LABORCOSTWAGE,ANDTHEDEGREEOFOPENNESSANDPROGRESSOFREFORMSHAREAGENERALPOOLEDREGRESSIONMODELISUSEDONTHESEVARIABLESANDISSPECIFIEDASEQUATION2WHERESUBSCRIPTIREFERSTOINDIVIDUALPROVINCESIISTHEI
