FDI和贸易对经济增长的影响：来自发展中国家的实证分析【外文翻译】.doc

1、1本科毕业论文外文翻译外文题目IMPACTOFFOREIGNDIRECTINVESTMENTANDTRADEONECONOMICGROWTHEVIDENCEFROMDEVELOPINGCOUNTRIES出处AMERICANJOURNALOFAGRICULTURALECONOMICS,VOL86ISSUE3作者MAKKI,SHIVAS,SOMWARU,AGAPI原文IMPACTOFFOREIGNDIRECTINVESTMENTANDTRADEONECONOMICGROWTHEVIDENCEFROMDEVELOPINGCOUNTRIESFOREIGNDIRECTINVESTMENTFDIANDTR

2、ADEAREOFTENSEENASIMPORTANTCATALYSTSFORECONOMICGROWTHINTHEDEVELOPINGCOUNTRIESFDIISANIMPORTANTVEHICLEOFTECHNOLOGYTRANSFERFROMDEVELOPEDCOUNTRIESTODEVELOPINGCOUNTRIESFDIALSOSTIMULATESDOMESTICINVESTMENTANDFACILITATESIMPROVEMENTSINHUMANCAPITALANDINSTITUTIONSINTHEHOSTCOUNTRIESINTERNATIONALTRADEISALSOKNOWNT

3、OBEANINSTRUMENTOFECONOMICGROWTHROMERTRADEFACILITATESMOREEFFICIENTPRODUCTIONOFGOODSANDSERVICESBYSHIFTINGPRODUCTIONTOCOUNTRIESTHATHAVECOMPARATIVEADVANTAGEINPRODUCINGTHEMEVENTHOUGHPASTSTUDIESSHOWTHATFDIANDTRADEHAVEAPOSITIVEIMPACTONECONOMICGROWTH,THESIZEOFSUCHIMPACTMAYVARYACROSSCOUNTRIESDEPENDINGONTHELE

4、VELOFHUMANCAPITAL,DOMESTICINVESTMENT,INFRASTRUCTURE,MACROECONOMICSTABILITY,ANDTRADEPOLICIESTHELITERATURECONTINUESTODEBATETHEROLEOFFDIANDTRADEINECONOMICGROWTHASWELLASTHEIMPORTANCEOFECONOMICANDINSTITUTIONALDEVELOPMENTSINFOSTERINGFDIANDTRADETHISARTICLEANALYZESTHEROLEOFFDIANDTRADEINPROMOTINGECONOMICGROW

5、THACROSSSELECTEDDEVELOPINGCOUNTRIESANDTHEINTERACTIONAMONGFDI,TRADE,ANDECONOMICGROWTHWEEXAMINEDATAFROMSIXTYSIXDEVELOPINGCOUNTRIESOVERTHELASTTHREEDECADESOURRESULTS2SUGGESTTHATFDI,TRADE,HUMANCAPITAL,ANDDOMESTICINVESTMENTAREIMPORTANTSOURCESOFECONOMICGROWTHFORDEVELOPINGCOUNTRIESWEFINDASTRONGPOSITIVEINTER

6、ACTIONBETWEENFDIANDTRADEINADVANCINGECONOMICGROWTHOURRESULTSALSOSHOWTHATFDISTIMULATESDOMESTICINVESTMENTTHECONTRIBUTIONOFFDITOECONOMICGROWTHISENHANCEDBYITSPOSITIVEINTERACTIONWITHHUMANCAPITALANDSOUNDMACROECONOMICPOLICIESMETHODOLOGYANDDATAOURECONOMETRICMODELISDERIVEDFROMAPRODUCTIONFUNCTIONINWHICHTHELEVE

7、LOFACOUNTRYSPRODUCTIVITYDEPENDSONFDI,TRADE,DOMESTICINVESTMENT,HUMANCAPITAL,ANDINITIALGROSSDOMESTICPRODUCTGDPPERCAPITATHEMODELISBASEDONENDOGENOUSGROWTHTHEORY,INTHETRADITIONOFBALASUBRAMANYAM,SALISU,ANDSAPSFORDANDBORENSZTEIN,GREGORIO,ANDLEE,WHEREFDICONTRIBUTESTOECONOMICGROWTHDIRECTLYTHROUGHNEWTECHNOLOG

8、IESANDOTHERINPUTSASWELLASINDIRECTLYTHROUGHIMPROVINGHUMANCAPITAL,INFRASTRUCTURE,ANDINSTITUTIONSTOASSESSEMPIRICALLYTHEEFFECTSOFFDIANDTRADEONECONOMICGROWTH,WESPECIFYTHEFOLLOWINGBASICFORMULATIONGAB1FDIB2TRDB3HCB4KB5G0C1FDITRDC2FDIHCC3FDIK（1）D1TRID2TXD3GCEWHEREGISTHEPERCAPITAGDPGROWTHRATETRD,THETRADEEXPO

9、RTSPLUSIMPORTSOFGOODSANDSERVICESHC,THESTOCKOFHUMANCAPITALK,THEDOMESTICCAPITALINVESTMENTG0,THEINITIALGDPINITIALSTOCKIRT,THEINFLATIONRATETX,THETAXONINCOME,PROFITS,ANDCAPITALGAINSINTHEHOSTCOUNTRYEXPRESSEDASPERCENTAGEOFCURRENTREVENUEANDGCISGOVERNMENTCONSUMPTIONTHEVARIABLESFDI,TRD,K,GCAREMEASUREDASRATIOS

10、TOGDPWEALSOACCOUNTFORINTERACTIONOFFDIWITHTRADEANDDOMESTICINVESTMENT,INADDITIONTOHUMANCAPITALPASTEMPIRICALSTUDIESHAVEINDICATEDTHATFDI,TRADE,HUMANCAPITAL,ANDDOMESTICINVESTMENTHAVEAPOSITIVEIMPACTONECONOMICGROWTHINDEVELOPINGCOUNTRIESWEEXPECTTHEESTIMATEDCOEFFICIENTSFORTHESEVARIABLESTOBEPOSITIVEWEALSOEXPE

11、CTPOSITIVEINTERACTIONSBETWEENFDIANDTRADEANDFDIANDDOMESTICCAPITALINVESTMENTINPROMOTINGECONOMICGROWTHTHESTOCKOFHUMANCAPITALINAHOSTCOUNTRYISCRITICALFORABSORBINGFOREIGNKNOWLEDGEANDANIMPORTANTDETERMINANTOFWHETHERPOTENTIALSPILLOVERSWILLBEREALIZEDWEPOSTULATENOTONLYAPOSITIVERELATIONSHIPBETWEENFDIANDTHEGDPGR

12、OWTHRATEBUTALSOAPOSITIVE3INTERACTIONBETWEENFDIANDHUMANCAPITALINADVANCINGECONOMICGROWTHTHEAPPLICATIONOFADVANCEDTECHNOLOGIESEMBODIEDINFDIREQUIRESASUFFICIENTLEVELOFHUMANCAPITALTHATIS,THEHIGHERTHELEVELOFHUMANCAPITALINAHOSTCOUNTRY,THEHIGHERTHEEFFECTOFFDIONTHECOUNTRYSECONOMICGROWTHONEOFTHEKEYQUESTIONSREGA

13、RDINGFDIANDECONOMICGROWTHISWHATISTHEINTERACTIONBETWEENFDIANDDOMESTICINVESTMENTASARGUEDBEFORE,FDIISANIMPORTANTVEHICLEFORTHETRANSFEROFCAPITAL,TECHNOLOGY,ANDKNOWLEDGETOHOSTCOUNTRIES,THEREBYGENERATINGHIGHGROWTHOPPORTUNITIESINPRACTICE,HOWEVER,THEGROWTHENHANCINGIMPACTOFFDIDEPENDSCRITICALLYONTHEABSORPTIVEC

14、APACITYOFAHOSTCOUNTRYANDWHETHERFDICROWDSOUTITSDOMESTICINVESTMENTTHUS,ANIMPORTANTQUESTIONTOBEADDRESSEDISWHATISTHEEXTENTTOWHICHFDISUBSTITUTESFORORCOMPLEMENTSDOMESTICINVESTMENTINOUREMPIRICALMODEL,WEINCLUDEFDIANDDOMESTICINVESTMENTSEPARATELYASWELLASANINTERACTIONTERMBETWEENFDIANDDOMESTICINVESTMENTFDIKAPOS

15、ITIVECOEFFICIENTFORTHEINTERACTIONTERMWOULDSUGGESTTHATFDIANDDOMESTICINVESTMENTKREINFORCECOMPLEMENTEACHOTHERINADVANCINGECONOMICGROWTHTHEINITIALGDP,MEASUREDINTERMSOFCONSTANTUSDOLLARS,CONTROLSFORPREEXISTINGECONOMICANDINSTITUTIONALCONDITIONSINTHEHOSTECONOMYWEEXPECTTHEINITIALGDPEXPRESSEDINLOGARITHMSTOBENE

16、GATIVELYRELATEDWITHGDPGROWTHRATESTHEINFLATIONRATEISAKEYINDICATOROFFISCALANDMONETARYPOLICIESOFACOUNTRYALOWERINFLATIONRATESHOULDMEANABETTERCLIMATEFORINVESTMENT,TRADE,AND,THEREFORE,ECONOMICGROWTHGOVERNMENTCONSUMPTIONANDTAXONINCOME,PROFITS,ANDCAPITALGAINSAREPROXIESFORINSTITUTIONSANDINFRASTRUCTUREINTHEHO

17、STCOUNTRIESSINCEOUROBJECTIVEISTOQUANTIFYTHEEFFECTSOFFDIANDTRADEONECONOMICGROWTH,WEFOCUSONDEVELOPINGCOUNTRIESDATAFOROURANALYSISAREOBTAINEDFROMTHEWORLDDEVELOPMENTINDICATORSWDIDATABASEHOWEVER,WELIMITOURANALYSISTO1971THROUGH2000BECAUSETHEFLOWOFFDITOMOSTDEVELOPINGCOUNTRIESBEGANIN1970SALLVARIABLESREPRESEN

18、TTHEAVERAGEOVERTHEFOLLOWINGDECADES19711980,19811990,AND19912000WEESTIMATEASYSTEMOFTHREEEQUATIONS,WHERETHEDEPENDENTVARIABLESARETHEMEANVALUESOFPERCAPITAGDPGROWTHRATESINEACHDECADEWEESTIMATETHESYSTEMOFEQUATIONSUSINGTHESEEMINGLYUNRELATEDREGRESSIONSURMETHODASWELLASINSTRUMENTALVARIABLETHREESTAGELEASTSQUARE

19、S,TSLSAPPROACHTHESURESTIMATIONALLOWSFORDIFFERENTERRORVARIANCESINEACHEQUATIONANDFORCORRELATIONOFTHESEERRORSACROSSEQUATIONS,WHILETHE4INSTRUMENTALVARIABLETECHNIQUEALLOWSUSTOOVERCOMEPOTENTIALBIASESINDUCEDBYENDOGENEITYPROBLEMSBETWEENFDIANDECONOMICGROWTHEMPIRICALRESULTSTHEPURPOSEOFOUREMPIRICALINVESTIGATIO

20、NISTOANALYZETHEEFFECTSOFFDIANDTRADEONECONOMICGROWTHANDTOEXAMINEHOWFDIINTERACTSWITHTRADE,HUMANCAPITAL,ANDDOMESTICINVESTMENTINADVANCINGECONOMICGROWTHINDEVELOPINGCOUNTRIESWECONTROLFORPREEXISTINGECONOMICCONDITIONSBYINCLUDINGINITIALGDPASONEOFTHEEXPLANATORYVARIABLESWEALSOACCOUNTFORDIFFERENCESINMACROECONOM

21、ICPOLICIESINTHEHOSTCOUNTRIESBYINCLUDINGVARIABLES,SUCHASINFLATIONRATE,TAXBURDEN,ANDGOVERNMENTCONSUMPTIONTABLE1PRESENTSTHEECONOMETRICRESULTSREGRESSIONS11,12,AND13,DIFFERENTVARIANTSOFEQUATION1ABOVEAREESTIMATEDUSINGTHESURMETHODREGRESSION11ISOURBASICSPECIFICATIONWITHEXPLANATORYVARIABLESOFFDI,TRADE,HUMANC

22、APITAL,DOMESTICINVESTMENT,ANDINITIALGDPREGRESSION12EXTENDS11TOINCLUDEINTERACTIONOFFDIWITHTRADE,HUMANCAPITAL,ANDDOMESTICINVESTMENTREGRESSION13BUILDSONREGRESSION12BYCONTROLLINGFORINFLATIONRATE,TAXBURDEN,ANDGOVERNMENTCONSUMPTIONOURRESULTSSHOWTHATMOSTCOEFFICIENTSHAVETHEEXPECTEDSIGNS,PARTICULARLYINSPECIF

23、ICATION13THEESTIMATEDR2AREGENERALLYLOWBUTREASONABLEGIVENTHECROSSSECTIONALNATUREOFTHEDATAUSEDREGRESSION11REVEALSTHATFDIANDTRADEHAVEAPOSITIVEIMPACTONECONOMICGROWTHAFTERCONTROLLINGFORHUMANCAPITAL,DOMESTICINVESTMENT,ANDINITIALINCOMETHEESTIMATEDCOEFFICIENTFORFDIISPOSITIVEANDSTATISTICALLYSIGNIFICANTWHILET

24、HEESTIMATEDCOEFFICIENTFORTRADEISNOTSTATISTICALLYSIGNIFICANTSINCETHECOEFFICIENTOFFDIISLARGERTHANTHECOEFFICIENTOFTRADE,ITINDICATESTHEDIFFERENTIALIMPACTOFFDIINTHEHOSTCOUNTRYSECONOMICGROWTHTHECOEFFICIENTFORHUMANCAPITALISPOSITIVE,IMPLYINGTHATHUMANCAPITALCONTRIBUTESPOSITIVELYTOECONOMICGROWTHSIGNIFICANTONL

25、YATACONFIDENCELEVELOF88THECOEFFICIENTSFORDOMESTICINVESTMENTANDINITIALINCOMEARENOTSTATISTICALLYSIGNIFICANTINCLUDINGINTERACTIONSBETWEENFDIANDTRADE,FDIANDHUMANCAPITAL,ANDFDIANDDOMESTICINVESTMENTNOTONLYIMPROVESTHEOVERALLPERFORMANCEOFTHEESTIMATIONBUTALSOALLOWSUSTO5CAPTURETHEIRINTERACTIONEFFECTSONECONOMIC

26、GROWTHINREGRESSION12,THEINTERACTIONOFFDIANDTRADEYIELDSAPOSITIVEANDSTATISTICALLYSIGNIFICANTCOEFFICIENTWHILETHEEFFECTSOFFDIANDTRADE,BYTHEMSELVES,AREPOSITIVEBUTNOTSTATISTICALLYSIGNIFICANTREGRESSION12ALSOREVEALSTHATTHEFDIINTERACTSPOSITIVELYWITHDOMESTICINVESTMENTINADVANCINGECONOMICGROWTHTHEESTIMATEDCOEFF

27、ICIENTFORDOMESTICINVESTMENTISPOSITIVEANDSTATISTICALLYSIGNIFICANTATACONFIDENCELEVELOF90THEESTIMATEDCOEFFICIENTSINDICATETHATHOSTCOUNTRIESBENEFITPOSITIVELYBOTHFROMFDI,ITSELF,ANDTHROUGHFDISPOSITIVEINTERACTIONWITHTRADEANDDOMESTICINVESTMENTTHEINTERACTIONBETWEENFDIANDHUMANCAPITAL,ALTHOUGHPOSITIVE,ISNOTSTAT

28、ISTICALLYSIGNIFICANTREGRESSION13INCLUDESADDITIONALVARIABLESTOCONTROLFORMACROECONOMICPOLICIESANDINSTITUTIONALSTABILITYTHATCOULDHAVEASIGNIFICANTIMPACTONFDIANDTRADEAND,THUS,ONECONOMICGROWTHRECENTLITERATUREINDICATESTHATFDIISGREATLYINFLUENCEDBYHOSTCOUNTRYPOLICIES,SUCHASMONETARY,FISCAL,ANDOPENMARKETPOLICI

29、ESWEINCLUDEINFLATIONRATES,TAXINCOME,ANDGOVERNMENTCONSUMPTIONTHERESULTSOFREGRESSION13REVEALTHATFDIANDTRADECONTRIBUTEPOSITIVELYTOECONOMICGROWTH,BUTTHE6ESTIMATEDCOEFFICIENTSARENOTSTATISTICALLYSIGNIFICANTTHESTOCKOFHUMANCAPITALANDDOMESTICINVESTMENT,ONTHEOTHERHAND,HASPOSITIVEANDSTATISTICALLYSIGNIFICANTCOE

30、FFICIENTSTHERESULTSALSOINDICATETHATFDIPOSITIVELYINTERACTSWITHTRADE,HUMANCAPITAL,ANDDOMESTICINVESTMENTBUTONLYFDITRADEINTERACTIONISSTATISTICALLYSIGNIFICANTTHISIMPLIESTHATFDIANDTRADECOMPLEMENTEACHOTHERINADVANCINGGROWTHRATEOFINCOMEINDEVELOPINGCOUNTRIESTHISRESULTISCONSISTENTWITHTHEIDEATHATFLOWOFADVANCEDT

31、ECHNOLOGYBROUGHTALONGBYFDICANINCREASETHEGROWTHRATEOFTHEHOSTECONOMYBYINTERACTINGWITHTHATCOUNTRYSTRADETHEDIVERSEEXPERIENCESFROMDEVELOPINGCOUNTRIESSUGGESTTHATFDIANDTRADE,BYTHEMSELVES,MAYNOTGUARANTEEECONOMICGROWTHACOUNTRYSECONOMICGROWTHISALSOAFFECTEDBYITSMACROECONOMICPOLICIESANDINSTITUTIONALSTABILITYSOU

32、NDMACROECONOMICPOLICIESANDINSTITUTIONALSTABILITYARENECESSARYPRECONDITIONSFORFDIDRIVENGROWTHTOMATERIALIZETHEESTIMATEDCOEFFICIENTSFORTHETHREEPOLICYVARIABLESINFLATIONRATE,GOVERNMENTCONSUMPTION,ANDTAXONINCOME,PROFITS,ANDCAPITALGAINSARENEGATIVEANDSTATISTICALLYSIGNIFICANTTHISIMPLIESTHATLOWERINGTHEINFLATIO

33、NRATE,TAXBURDEN,ANDGOVERNMENTCONSUMPTIONWOULDPROMOTEECONOMICGROWTHLOWERINFLATIONRATESWOULDINDICATETHATTHEHOSTCOUNTRYSMACROECONOMICPOLICIESARESTABLEANDDISCIPLINEDLOWERTAXBURDENWOULDMAKETHEINVESTMENTS,FOREIGNANDDOMESTIC,MOREPROFITABLEDECREASINGTHEGOVERNMENTCONSUMPTIONWOULDLEAVEMOREMONEYFORINVESTMENTSF

34、DICROWDSINDOMESTICINVESTMENTONEOFTHEIMPORTANTQUESTIONSRAISEDINTHELITERATUREISWHETHERFDIAUGMENTSAHOSTCOUNTRYSCAPITALINVESTMENTORCROWDSOUTDOMESTICINVESTMENTEVENTHOUGHNOTSTATISTICALLYSIGNIFICANT,THEPOSITIVEINTERACTIONBETWEENFDIANDDOMESTICINVESTMENTINREGRESSION13IMPLIESTHATDOMESTICINVESTMENTISUNLIKELYTO

35、BECROWDEDOUTINDEVELOPINGCOUNTRIESTOFURTHERSTRENGTHENOURARGUMENT,WEESTIMATETHECONTRIBUTIONOFFDITODOMESTICINVESTMENTAFTERCONTROLLINGFORTRADE,HUMANCAPITAL,INITIALINCOMELEVELS,ANDVARIOUSMACROECONOMICPOLICYVARIABLESREGRESSION14INTABLE1PRESENTSTHERESULTSOFTHISESTIMATIONUSINGTHESURMETHODTHERESULTSINDICATEF

36、DIHASAPOSITIVEEFFECTONDOMESTICINVESTMENT,ASTHEESTIMATEDCOEFFICIENTISPOSITIVEANDSTATISTICALLYSIGNIFICANTTHISPOSITIVERELATIONSHIPIMPLIESTHATFDISTIMULATESORCROWDSINDOMESTICINVESTMENTTHISFINDINGISCONSISTENTWITHBORENSZTEIN,GREGORIO,ANDLEEEVENTHOUGHTRADE,BYITSELF,ISNOT7STATISTICALLYSIGNIFICANT,TRADEINTERA

37、CTSPOSITIVELYWITHFDIONDOMESTICINVESTMENTTHEESTIMATEDCOEFFICIENTFORTHEFDITRADEINTERACTIONTERMISPOSITIVEANDSIGNIFICANTATTHE90CONFIDENCELEVELENDOGENEITYPROBLEMSTHECORRELATIONBETWEENFDIANDGROWTHRATECOULDARISEFROMANENDOGENOUSDETERMINATIONOFFDITHATIS,FDI,ITSELF,MAYBEINFLUENCEDBYINNOVATIONSINTHESTOCHASTICP

38、ROCESSGOVERNINGGROWTHRATESBORENSZTEIN,GREGORIO,ANDLEEFOREXAMPLE,MARKETREFORMSINHOSTCOUNTRIESCOULDINCREASEBOTHGDPGROWTHRATESANDTHEINFLOWOFFDISIMULTANEOUSLYINTHISCASE,THEPRESENCEOFCORRELATIONBETWEENFDIANDTHECOUNTRYSPECIFICERRORTERMWOULDBIASTHEESTIMATEDCOEFFICIENTSTHEENDOGENEITYPROBLEMISADDRESSEDBYUSIN

39、GTHEINSTRUMENTALVARIABLESONEOFTHEMAJORPROBLEMSWITHTHEINSTRUMENTALVARIABLEESTIMATIONMETHODISTHEDIFFICULTYINIDENTIFYINGINSTRUMENTSTHATAREHIGHLYCORRELATEDWITHFDIORTRADEBUTNOTWITHTHEERRORTERMWEUSELAGGEDVALUESOFFDI,LAGGEDVALUESOFTRADE,ANDLOGVALUEOFTOTALGDPASINSTRUMENTSINATSLSMETHODTHERESULTSOFTHETSLSMODE

40、LREPORTEDINTABLE2,REGRESSIONS2123SHOWTHATTHEINSTRUMENTALVARIABLEESTIMATIONYIELDSQUALITATIVELYSIMILARRESULTSASTHOSEOBTAINEDBYTHESURMETHODTHEESTIMATEDCOEFFICIENTSONFDIANDTRADE,BYTHEMSELVES,AREPOSITIVEBUTSTATISTICALLYINSIGNIFICANTTHEINTERACTIVETERMOFFDIANDTRADEISPOSITIVEANDSTATISTICALLYSIGNIFICANTTHISA

41、LTERNATIVEESTIMATIONALSOSUGGESTSTHATOURRESULTSAREROBUST89本科毕业论文外文翻译外文题目IMPACTOFFOREIGNDIRECTINVESTMENTANDTRADEONECONOMICGROWTHEVIDENCEFROMDEVELOPINGCOUNTRIES出处AMERICANJOURNALOFAGRICULTURALECONOMICS,VOL86ISSUE3作者MAKKI,SHIVAS,SOMWARU,AGAPI译文FDI和贸易对经济增长的影响来自发展中国家的实证分析引言在发展中国家FDI和贸易常常被看作刺激经济增长的重要因素。FDI是

42、技术从发达国家传递到发展中国家的载体。FDI还可以刺激东道国的国内投资，提高人力资本和制度的便利。国际贸易也一直是促进经济增长的工具（罗默）。贸易使得具有比较优势的国家生产的产品和服务更便利的在国家间流通。尽管过去的研究证明FDI和对外贸易对经济增长有积极正相关的影响，但是这种影响的大小从国家整体来说取决于人力资本的水平、国内投资、基础设施、宏观经济稳定性和贸易政策。本文主要延续之前的讨论关于FDI和贸易对经济增长的作用，经济和制度的发展在刺激FDI与贸易上的重要性。本文分析了FDI和贸易在经济增长中扮演的角色，通过选择66个发展中国家在最近三十年的数据分析FDI、贸易、经济增长之间的关系。实

43、证研究结果表明FDI、贸易、人力资本、国内投资是发展中国家经济增长的重要因素。结果还表明在推进经济增长方面FDI与投资之间存在积极正相关。同样，FDI刺激国内投资，但是FDI对经济增长的贡献可以由通FDI存在正相关性的人力资本、稳定的宏观经济政策来提高。二、研究方法和数据本文采用的经济模型来源于生产函数，一个国家的生产力水平取决于FDI、贸易、10国内投资、人力资本、最初的人均GDP。这个模型建立在内生增长理论上，在传统上BALASUBRAMANYAM,SALISU,ANDSAPSFORDANDBORENSZTEIN,GREGORIO,ANDLEE的观点，FDI对经济增长的影响直接通过新技术和

44、其他要素投入或者通过人力资本、基础设施、制度的提高。为了实证分析FDI和贸易对经济增长的作用，本文建立如下的方程GAB1FDIB2TRDB3HCB4KB5G0C1FDITRDC2FDIHCC3FDIK（1）D1TRID2TXD3GCEG代表人均GDP增长率；TRD代表货物和服务的贸易（出口和进口）；HC表示人力资本存货；K表示国内资本投资；G0表示初期GDP；IRT表示通货膨胀率；TX是以当前收入的百分数，表示东道国的税收收入、利润、资本收益。GC表示政府消费。变量FDI、TRD、K、GC以比率表示GDP。本文还说明了FDI对贸易、国内投资、人力资本的关系。以前的实证分析证明国内投资对发展中国

45、家的经济增长有积极影响。我们预测这些变量的估计系数是正数，同样预测FDI和贸易、FDI和国内投资在促进经济增长方面有积极正相关作用。东道国的人力资本存量对吸收国外知识和实现潜在的FDI技术外溢至关重要。我们假定不仅FDI和GDP增长率存在积极正相关，而且在促进经济增长方面FDI和人力资本存在正互相关系。由FDI带来的先进技术的运用需要足够的人力资本，即东道国的人力资本水平越高，FDI对经济增长带来的效应越高。关于FDI和经济增长之间的关系有一个重要问题FDI和国内投资存在什么关系一直以来，FDI被作为资本、技术、知识转移到东道国的重要载体，因此带来刺激经济快速增长的机会。然而，实际上，FDI刺

46、激增长的作用取决于东道国吸收资本的能力，还存在FDI是否“挤出”国内投资的问题。因此，这个问题变成了FDI在多大程度上替代或补充国内投资在本文的实证模型中将FDI和国内投资分开，又将FDI和国内投资作为术语交汇（FDIK）来研究。FDIK的相关系数是正的话说明FDI和国内投资在促进经济增长时相互补充或加强。为控制东道国的现存的经济和制度条件以不变价格计量的初期GDP。我们预期以对数形式表示的初期GDP对GDP增长率表现负相关。通货膨胀率是一个国家财政和货币政策的重要指标。低的通货膨胀率意味着好的投资、贸易环境，因此有利于经济增长。政府消费和收入、利润、资本收益的税收代表东道国的制度和基础设施。

47、因为我们的目的是量化FDI和贸易对经济增长的效应，本文集中研究发展中国家。本文的数据来源于世界发展指标数据库（WDI）。本文之所以选择66个发展中国11家1971年到2000的数据时因为大部分发展中国家的FDI流入始于1970S。所有的变量的数据都是以下三个阶段19711980,19811990,19912000。本文运用三个等式组成的系统，因变量是每十年的人均GDP增长率。我们运用无相关回归（SUR）及工具变量（三步最小二乘法）来估计等式系统模型。SUR方法考虑到三个方程的不同误差变量及误差变量与等式的相关性，工具变量可以允许克服FDI和经济增长之间的内生性问题引起的潜在偏差。三、实证检验结

48、果本文的目的是实证分析FDI和贸易对经济增长的效应，检验发展中国家FDI如何通过贸易、人力资本、国内投资在经济增长中发挥作用。把初期GDP作为其中一个解释变量控制现存经济条件。通过加入通货膨胀率、税收负担、政府消费三个变量考虑东道国不同宏观经济政策。表1表示实证分析的结果，回归11,12,13是等式（1）用SUR方法估计的变形。回归11是解释变量为FDI、贸易、人力资本、国内投资、初期GDP的基本说明。回归12扩充了回归11，把FDI对贸易、人力资本、国内投资、初期GDP的关系包括在内。回归13通过控制通货膨胀率、税收负担、政府消费建立在回归12的基础上。实证分析的结果表明大多数回归系数符合预

49、期，特别是回归13的说明。可决系数R2虽比一般低但是对截面数据的分析是合理的。回归11说明了FDI和贸易在控制人力资本、国内投资、初期收入后对经济增长有正相关性。FDI的估计系数是正的并且在统计上显著，但是贸易的估计系数在统计上不显著。因为FDI的系数比贸易的系数大，说明两者对经济增长的不同作用。人力资本的系数是正的，说明人力资本对经济增长正相关（只在置信水平88下）。国内投资和初期收入的系数在统计上不显著。把FDI和贸易、FDI和人力资本、FDI和国内投资的关系包括在内不仅在整体上使估计偏高，还使得我们采集到他们对经济增长的影响。在回归2中，FDITRD的系数是正的而且统计上显著，但是就FDI和贸易单独来说，系数虽是正数但统计上不显著。回归2还说明在促进经济增长过程中FDI对国内投资正相关。在置信水平90下，国内投资的估计系数是正数，统计上显著。这表明东道国不仅受益于FDI本身还受益于FDI对贸易、国内投资的正相关作用。FDI和人力资本之间虽正相关但在统计上不显著。回归3为了控制宏观经济政策和制度稳定性加了另一些变量，这些变量对FDI、贸易甚至经济增长都有重要影响。最近的研究表明东道国的货币政策、财政政策、开放政策都对FDI影响很大。所以在回归3中加12入通货膨胀率、税收、政府消费。回归3的结果表明FDI和贸易促进经济