搜索
1FOREIGNDIRECTINVESTMENT,TRADEANDECONOMICGROWTH:AMULTI-COUNTRYEVIDENCESong-ZanChiouWeiGraduateInstituteofEconomicsNanHuaUniversityChiou-MingHuangGraduateInstituteofEconomicsNanHuaUniversity2inwardFDI()GrangerstructuralVAR19652000VARABSTRACTAlthoughthereisconsiderableempiricalevidenceonthelinkbetweenforeigndirectinvestment(FDI)andeconomicgrowthortradeandeconomicgrowthinthedevelopingcountries,casualpatternsamongthesevariableshavenotbeenyetfullyexplored.Thispaperconstructsastructuralvectorautoregressionmodel(structuralVAR)tostudytheinteractionsamongfourimportantmacroeconomicvariables,whichareFDI,export,importandGDPgrowth,withobservationsthrough1965-2000fromnineeconomiesinEastAsiaandLatinAmerica.Themainresultscanbesummarizedasfollow:First,wefindthatexportexpansionisthemaindeterminantsofinwardFDI.Second,inLatinAmerica,export,importandGDPinfluenceFDIbutFDIdoesn’thaveasignificantimpactontheothervariables.Third,inEastAsia,FDIhaveapositiveeffectonGDPandexport.Atlast,lowwagelevelandhighlaborsupplyarenotthesignificantdeterminantsofinwardFDI.Keywords:inwardFDI,EconomicGrowth,StructuralVAR,InnovationAccounting3multinationalcorporationsMNCinwardforeigndirectinvestmentFDI,1Borenszteinetal.1998outputhomecountryinputChenetal.1995Chen1996BroadmanandSalisu1996Rueberetal.1973KhanandLeng19974Dollar(1992)Marin1992Edward(1993)Enders1995Granger1969Sim1980Riezman,etal.1996Granger1969Ibrahim2000Granger1969JunandSingh1996KhanandLeng1997Luiz.andKilchiro.(2000)Zhang2001BlanchardandQuah1989VARstructuralVARVAR51965-200011960199919601960-196979%1970-197980%1986-199089%199091%19861986Braga199219904001960196570%1(1960-2000)1960-19691970-19791980-19851986-19901991-19951996-2000FDI()2.25132360198FDI(%)13214221611FDI()0.4151435101FDI(%)162038545951FDI(%)11167282110FDI()1.43691880FDI(%)636046352940FDI(%)1520-5171918119992001WorldInvestmentReport/UNCTAD2BalanceofPaymentsStatisticalYearbook/IMF621988-1990198860%20~30%199761.5%29.2%32.8%67.3%38.8%23.8%55.5%1990219881997/6832910039379100904731008399510057308.434808.8309583.448172.74219261.72651867.355558761.53254938.81998329.2936223.829628032.84660755.54240.7190.1206482.3220121999WorldInvestmentReport/UNCTAD7(2000)131988-19901998-200031988-19901998-20001988-19901998-20001/2/3/41/2/3/41.00.20.70.61.00.30.70.750.80.61.00.81.11.01.61.261.30.20.70.71.10.20.60.6200112LABSTA200134568419941994-1995199519974FDIGDP(1985-1999)1985-19951996199719981999FDIGDP%FDIGDP%FDIGDP%FDIGDP%FDIGDP%2.27.06.513.29.216.17.312.224.147.71.82.210.57.018.711.728.518.431.431.30.77.73.114.85.626.24.224.44.038.64.59.29.215.512.816.411.613.211.911.70.90.92.31.23.11.85.45.710.69.32.917.17.317.06.515.12.713.93.520.14.129.39.025.68.122.16.320.67.226.11.02.71.93.02.23.40.20.42.94.41.44.12.43.23.77.85.111.63.615.41219992001WorldInvestmentReport/UNCTAD3BalanceofPaymentsStatisticalYearbook/IMF9---Borenszteinetal.1998Balasubramanyametal.(1996)(capitalstock)know-howtechnologymarket-seekingFDIMarkusenetal.1996growth-drivenFDIWheelerandMody1992Zhang2000---Karikari19921961-198810DeMello1997OECDZhang2001Granger2115---Chen1996BroadmanandSalisu1996Rueberetal.1973Caves,(1996)Helleiner,(1989)ZhangandSong200011firm-specificadvantages---JunandSingh19961969-1993Luiz.andKilchiro.(2000)(VECM)Granger(1987)Liu,WangandWei2001Granger1984-199819Zhang,andFelmingham,2001464312KhanandLeng1997Engle-Granger1987Johansen19881990Granger19871965-1995KhanandLeng1997Sim(1980)(reducedformstructuralvectorauto-regressionmodel)BlanchardandQuah(1989)tY(n×1)tYtjtpijtAYBYυ+=−=∑0(1)tυ()'ttυυΕ=υ∑tυjΒn×nAn×n(nonsingularmatrix)(1)1ptjtjtjYCYµ−==+∑(2)13jC=()10jIBB−−tµ()'ttEµµµ=∑0tttBAµµυ=+(3)(3)(contemporaneous)(2)tµ(3)(3)υ∑µ∑Α0Β)()()(1001′′−Σ−=Σ−−ABIBIAµν(4)0Βµ∑n(n+1)/2(4)BlanchardandQuah1989(0Β)A4×46(FDI)(FDI)61112223334441000011111101101ttttttttttttµµυµµυµµυµµυ+(5)itµitυi=1,2,3,4141965-200019651965TaiwanStatisticalDataBookInternationalMonetaryFund(IMF)InternationalFinancialStatistics(IFS)(EX)(1995=100)(IM)(1995=100)GDP(1995=100)inwardFDI(1995=100)VAR(overparameterization)AICAkaikeInformationCriterionSBCSchwarzInformationCriterion4(LikelihoodRatioTest)5LujingBoxQwhitenoiseJarque-BeranormalityQJarque-Bera15%115DFDickey-Fuller19791981ADFAugmentedDickeyFuller1984PPPhillipsandPerron1988PPPPseriallycorrelated(heterogeneity)VAR5VAR5FDIEXGDPIM-3.47-5.54*-3.31-5.56*-4.22*-7.31*-2.97-5.29*-1.54-6.40*-6.72*-10.33*-2.90-7.45*-2.05-4.74*-2.85-12.59*-2.82-22.21*-3.77*-27.38*-2.04-18.01*-3.66*-7.14*-2.13-3.72*-1.55-5.67*-2.93-4.41*-1.63-4.79*-2.87-5.02*-2.61-5.86*-2.32-4.02*-4.21*-9.63*-1.38-5.03*0.61-3.29-2.08-4.38*-2.26-6.36*-1.99-4.52*-0.734.26*-1.25-3.86*-4.06*-8.46*-1.60-4.91*-0.91-5.01*-2.01-5.24*-3.48-6.16*-2.52-4.58*-1.20-3.69*-2.18-4.17*1PP5%-3.542*16()GrangerrepresentationtheoremEngleandYoo(1987)Johansen1988JohansenJuselius1990Johansen6PP(1988)I(1)JohansenttttktttDXXXHεµφ++Π+∆Γ=∆−−−=∑11111:7βα′=Π:2Htracetestmaximaleigenvaluetest()()∑−=−−=−priiTHHQ1121ln|;ln2λ8()rrankH≤Π:0()()()()irTrHrHQ+−−=+−λ1ln1|;ln2129()rrankH=Π:0Johansenlikelihood