1、本科毕业论文外文翻译外文题目DODOMESTICCHINESEFIRMSBENEFITFROMFOREIGNDIRECTINVESTMENT出处UNIVERSITYOFCALIFORNIA,BERKELEY作者CHANGTAIHSIEH原文SUMMARYTHISPAPERUSESAUNIQUEPLANTLEVELPANELDATASETFROMTHECHINESEMANUFACTURINGSECTORTOMEASURETHEEFFECTSOFFOREIGNFIRMSONTHEPRODUCTIVITYOFDOMESTICFIRMSWEFINDTHATFOREIGNFIRMSAREMOREPROD
2、UCTIVETHANDOMESTICFIRMSINTHESAMEINDUSTRYWEFINDLITTLEEVIDENCETHATFOREIGNFIRMSHADANYEFFECT,EITHERPOSITIVEORNEGATIVE,ONTHEAVERAGEPRODUCTIVITYOFDOMESTICCHINESEMANUFACTURINGPLANTSHOWEVER,WEALSOFINDCLEAREVIDENCETHATTHISAVERAGEEFFECTMASKSIMPORTANTHETEROGENEITYLARGEDOMESTICFIRMSAPPEARTOBENEFITSIGNIFICANTLYF
3、ROMTHEPRESENCEOFFOREIGNFIRMSINTHESECTOR,WHILETHEPRODUCTIVITYOFSMALLDOMESTICFIRMSISLOWEREDBYTHEPRESENCEOFFOREIGNFIRMS1INTRODUCTIONINRECENTDECADES,DIRECTFOREIGNINVESTMENTDFIHASEXPLODEDINDEVELOPINGCOUNTRIESBYSOMEACCOUNTS,DFIACCOUNTSFORMORETHANHALFOFALLEXTERNALFINANCINGINDEVELOPINGCOUNTRIESINFACT,MANYPO
4、LICYMAKERSBELIEVETHATATTRACTINGDFIISACRUCIALINGREDIENTFORACCELERATINGTHECOUNTRYSECONOMICDEVELOPMENTFOREIGNFIRMS,THEARGUMENTGOES,WOULDINTRODUCENEWTECHNOLOGIESANDHELPUPGRADETECHNOLOGICALCAPABILITIESOFDOMESTICFIRMSTHISCOULDOCCURTHROUGHAVARIETYOFCHANNELSFOREXAMPLE,WORKERSEMPLOYEDBYFOREIGNFIRMSCOULDLEARN
5、THENEWTECHNOLOGIESANDDIFFUSETHESETECHNOLOGIESBYSTARTINGTHEIROWNFIRMSORBYMOVINGTOOTHERDOMESTICFIRMSITMIGHTBETHECASETHATFOREIGNFIRMSHELPUPGRADETHETECHNOLOGIESOFDOMESTICFIRMSBYSTIMULATINGUPSTREAMANDDOWNSTREAMLINKAGESDUETOTHESEREASONS,MANYCOUNTRIESHAVEINTRODUCEDABATTERYOFTAXINCENTIVES,RANGINGFROMTAXHOLI
6、DAYSTOTHEPROVISIONOFUTILITIESANDOTHERINFRASTRUCTUREATCONCESSIONARYRATESANDNOWHEREISTHISMOREAPPARENTTHANINCHINA,WHERETHEGOVERNMENTHASINTRODUCEDANUMBEROFINCENTIVESTOATTRACTFOREIGNINVESTMENT,WITHREMARKABLESUCCESSSOFARINFACT,CHINAHASEMERGEDASTHELARGESTRECIPIENTSOFDIRECTFOREIGNINVESTMENTINTHEWORLDYET,THE
7、REISREMARKABLYLITTLEEVIDENCEONTHEEFFECTOFFOREIGNFIRMSONTHETECHNOLOGICALCAPABILITIESOFDOMESTICFIRMSINCHINAARECENTIMPORTANTPAPERBYAITKENANDHARRISON1999USESSOPHISTICATEDECONOMETRICTECHNIQUESTOMEASURETHEEFFECTOFFOREIGNFIRMSONTHEPRODUCTIVITYOFDOMESTICFIRMSUSINGAPLANTLEVELCENSUSOFFIRMSINVENEZUELAHERE,AITK
8、ENANDHARRISON1999FINDLITTLEEVIDENCETHATDFIHADAPOSITIVEEFFECTONTHEPRODUCTIVITYOFDOMESTICFIRMSTHISPAPERWILLFOLLOWTHEAPPROACHTAKENBYAITKENANDHARRISONAPPLIEDTOAUNIQUEPANELDATASETOFCHINESEFIRMSSPECIFICALLY,THEREARETWOQUESTIONSWESEEKTOANSWERFIRST,DOJOINTVENTURESORWHOLLYOWNEDFOREIGNSUBSIDIARIESINCHINAHAVEH
9、IGHERLEVELSOFPRODUCTIVITYTHANDOMESTICFIRMSINSIMILARSECTORSSECOND,DOESTHEPRESENCEOFFOREIGNFIRMSEITHERJOINTVENTURESORWHOLLYOWNEDFOREIGNFIRMSRESULTINATECHNOLOGY“SPILLOVERS”TODOMESTICFIRMSWEREPORTTHREEMAINFINDINGSFIRST,FOREIGNFIRMSAREINDEEDMOREPRODUCTIVETHANDOMESTICFIRMSINTHESAMESECTORSECOND,THEPRESENCE
10、OFFOREIGNFIRMSINASECTORDOESNOTAPPEARTOHAVEANEFFECT,EITHERPOSITIVEORNEGATIVE,ONTHEPRODUCTIVITYOFANAVERAGEPLANTINTHESECTORTHIRD,FOREIGNPENETRATIONINASECTORHASIMPORTANTEFFECTSONINEQUALITYLARGEDOMESTICPLANTSAPPEARTOBEBENEFITFROMTHEPRESENCEOFFOREIGNFIRMS,WHILESMALLDOMESTICPLANTSAPPEARTOSUFFERTHISPAPERCON
11、TRIBUTESTOTHELARGELITERATUREONTHEEFFECTSOFFOREIGNINVESTMENTONTHEPRODUCTIVITYOFDOMESTICFIRMSINADDITIONTOTHEPAPERSCITEDABOVE,OTHERREPRESENTATIVEPAPERSFROMTHISLITERATUREINCLUDEBLOMSTROM1986FORMEXICANMANUFACTURINGANDGLOBERMAN1979FORCANADAASFORCHINA,WEARENOTAWAREOFANYPAPERTHATUSESFIRMLEVELDATATOEXAMINETH
12、EEFFECTOFFOREIGNINVESTMENTONDOMESTICPRODUCTIVITYTHEREFORE,THEMAINCONTRIBUTIONOFTHEPAPERISTOPROVIDETHEFIRSTEVIDENCEOFTHEEFFECTSOFFOREIGNCAPITALINACOUNTRYTHATISTHELARGESTHOSTOFFOREIGNDIRECTINVESTMENTINTHEWORLDTHEPAPERPROCEEDSASFOLLOWSSECTION2DESCRIBESTHEDATATHATISUSEDFORTHEANALYSISSECTION3PRESENTSTHEM
13、AINEMPIRICALRESULTSEXAMININGTHERELATIONSHIPBETWEENFOREIGNINVESTMENTANDPRODUCTIVITYATTHEPLANTLEVELSECTION4EXAMINESWHETHERTHESPILLOVERSARELOCALINNATURESECTION5LOOKSFOREVIDENCETHATFOREIGNCAPITALHADADIFFERENTIALEFFECTFORLARGEVERSUSSMALLFIRMSSECTION6CONCLUDES2DATATHEDATAUSEDINTHISPAPERISTHEFIRMLEVELDATAF
14、ROMTHECHINESEANNUALSURVEYOFINDUSTRIESCONDUCTEDANNUALLYBYCHINASNATIONALBUREAUOFSTATISTICSTHISISTHEDATAPUBLISHEDINTHEINDUSTRIALSTATISTICSCHAPTEROFTHEANNUALPUBLICATIONOFCHINASSTATISTICALYEARBOOKWEHAVETHISDATAFROM1998THROUGH2004THISSURVEYCOVERSALLSTATEPLANTSANDALLNONSTATEPLANTSWITHREVENUESGREATERTHAN5MI
15、LLIONYUANTHENUMBEROFPLANTSRANGESFROM200,000TO250,000THEDATAALSOPROVIDESPLANTIDENTIFIERS,SOWEAREABLETOTRACKPLANTSOVERTIMEFOREACHPLANT,THEDATASETALSOPROVIDESINFORMATIONONTHEPLANTSINDUSTRY,GEOGRAPHICLOCATION,OWNERSHIP,BOOKVALUEOFTHEPLANTSCAPITALSTOCK,INPUTCOST,REVENUES,EMPLOYMENT,ANDREVENUESTHEPIECESOF
16、INFORMATIONWEUSEFROMTHISDATASETARETHEFOLLOWINGFIRST,WEDEFINETHEPLANTSINDUSTRYASA3DIGITLEVELTHEREAREROUGHLY1803DIGITINDUSTRIESINTHEDATASECOND,WEUSETHEPLANTSREVENUES,EMPLOYMENT,INPUTCOSTS,ANDTHEBOOKVALUEOFTHECAPITALSTOCKTOMEASURETHEPLANTSPRODUCTIVITYTHEUSEOFTHEBOOKVALUEOFTHECAPITALSTOCKINSTEADOFTHEMAR
17、KETVALUEMAYBEOFCONCERNONEWAYINWHICHWETRIEDTODEALWITHTHISISTOUSETHEDATAONTHEAGEOFTHEFIRMTOCONSTRUCTAPROXYFORTHEMARKETVALUEOFCAPITALUSINGCONSTANTGROWTHRATEASSUMPTIONSANDTHEDATAONTHEBOOKVALUEOFTHECAPITALSTOCKFORTHERESULTSWEPRESENTBELOW,ITDOESNOTMAKEADIFFERENCEWHETHERWEUSETHEBOOKVALUEOFTHECAPITALSTOCKRE
18、PORTEDINTHEDATASETORWHETHERWEUSEOURPROXYFORTHEMARKETVALUEOFTHECAPITALSTOCKBUTWECANOBVIOUSLYPROVIDETHESERESULTSUPONREQUESTTHIRD,AKEYPIECEOFINFORMATIONPROVIDEDINTHEDATAISTHEOWNERSHIPOFTHEPLANTSPECIFICALLY,THEDATASETPROVIDESINFORMATIONONTHEOWNERSHIPOFTHEPLANTDIVIDEDINTOSTATE,LOCALGOVERNMENTS,COOPERATIV
19、E,ANDFOREIGNOWNERSHIPWEUSETHISINFORMATIONTOCONSTRUCTTHREEVARIABLESFIRST,WECONSTRUCTAVARIABLEMEASURINGTHEFOREIGNOWNERSHIPSHAREOFTHEPLANTWECALLTHISVARIABLEPLANT_DFISECOND,WECONSTRUCTAVARIABLEMEASURINGTHESHAREOFFOREIGNFIRMSINEACH3DIGITINDUSTRYSPECIFICALLY,WEMEASURETHEFOREIGNSHAREINANINDUSTRYASAWEIGHTED
20、AVERAGEOFTHEEMPLOYMENTSHAREOFEACHPLANTWHERETHEWEIGHTSARETHESHAREOFFOREIGNEQUITYINEACHPLANTWECALLTHISVARIABLESECTOR_DFIFINALLY,WECONSTRUCTAVARIABLEFORTHEIMPORTANCEOFFOREIGNFIRMSINAGEOGRAPHICREGIONSPECIFICALLY,WEDEFINETHELOCALREGIONASACHINESEPROVINCE,ANDMEASURETHESHAREOFFOREIGNFIRMSASAWEIGHTEDAVERAGEO
21、FTHEEMPLOYMENTSHAREOFEACHPLANTINTHEPROVINCE,WHERETHEWEIGHTSARETHEFOREIGNEQUITYSHAREOFEACHPLANTWECALLTHISVARIABLELOCAL_SECTOR_DFI3DODOMESTICFIRMSBENEFITFROMFOREIGNFIRMSTOESTIMATETHEEFFECTOFFOREIGNFIRMS,WEFOLLOWAITKENANDHARRISON1999ANDESTIMATETHEFOLLOWINGLOGLINEARPRODUCTIONFUNCTIONHERE,IINDEXESTHEPLAN
22、T,JINDEXESTHESECTOR,ANDTREPRESENTSTIMEASFORTHEVARIABLES,YISGROSSOUTPUTOFTHEFIRM,PLANT_DFIREPRESENTSTHESHAREOFFOREIGNEQUITYFORTHEPLANT,SECTOR_DFIREPRESENTSTHEEMPLOYMENTSHAREOFFOREIGNEQUITYINTHESECTOR,ANDCONTROLSDENOTESAVECTOROFCONTROLSAVECTOROFTIMEDUMMIES,THECOSTOFINTERMEDIATEINPUTS,THEEMPLOYMENTOFTH
23、EPLANT,ANDTHEBOOKVALUEOFTHECAPITALSTOCKTABLE1PRESENTSTHEBASICRESULTSFROMESTIMATINGEQUATION11ONTHECHINESEDATATHEDEPENDENTVARIABLEISTHELOGOFPLANTGROSSOUTPUT,WHICHISREGRESSEDONTHEMEASURESOFFOREIGNPARTICIPATION,THENUMBEROFWORKERSEMPLOYEDINTHEFIRM,THEVALUEOFINTERMEDIATEINPUTSUSEDBYTHEFIRM,ANDTHEBOOKVALUE
24、OFTHECAPITALSTOCKTHELASTTHREEVARIABLESAREINTRODUCEDINLOGSINADDITION,INALLTHEREGRESSIONSWEINCLUDEAVECTOROFTIMEDUMMIESINTHEFIRSTCOLUMN,WEALSOINCLUDEAVECTOROF3DIGITINDUSTRYDUMMIES187INDUSTRIESINTOTALTOALLOWFORFIXEDPRODUCTIVITYDIFFERENCESACROSSSECTORSASCANBESEEN,THECOEFFICIENTONTHEFOREIGNOWNERSHIPOFTHEP
25、LANTPLANT_DFIISPOSITIVEANDSTATISTICALLYSIGNIFICANTTHEPOINTESTIMATESUGGESTSTHATTHEPRODUCTIVITYOFPLANTTHATISENTIRELYFOREIGNOWNEDIS13PERCENTAGEPOINTSHIGHERTHANTHEPRODUCTIVITYOFAPLANTTHATISENTIRELYOWNEDBYDOMESTICINVESTORSINCONTRAST,THECOEFFICIENTONTHEFOREIGNOWNERSHIPINTHESECTORSECTOR_DFIDOESNOTAPPEARTOB
26、ESTATISTICALLYDIFFERENTFROMZEROTHISRESULTSUGGESTSTHATFOREIGNPRESENCEINASECTORASOPPOSEDTOBEINGFOREIGNOWNEDITSELFDOESNOTAPPEARTOHAVEANYEFFECTONTHEPRODUCTIVITYOFDOMESTICFIRMSTHECOEFFICIENTONTHETERMINTERACTINGFOREIGNOWNERSHIPOFTHEPLANTANDFOREIGNOWNERSHIPINTHESECTORPLANT_DFISECTOR_DFIISNEGATIVEANDSTATIST
27、ICALLYSIGNIFICANTTHENEGATIVECOEFFICIENTSUGGESTSTHATFORDOMESTICPLANTS,THEREAREPOSITIVESPILLOVERSFROMFOREIGNINVESTMENT,INCONTRASTTOFOREIGNPLANTSNOTETHATTHISRESULTDIFFERSFROMWHATAITKENANDHARRISON1999FOUNDINTHEIRANALYSISOFVENEZUELANMANUFACTURINGPLANTSTHESECONDCOLUMNPRESENTSTHECOEFFICIENTSFROMTHESAMEREGR
28、ESSION,BUTOMITSTHEINDUSTRYDUMMIESTHEREASONTHISISDONEISSOTHATWECANCOMPAREOURRESULTSWITHSTUDIESTHATUSEONLYAGGREGATEDINDUSTRYDATAASCANBESEEN,THECOEFFICIENTONFOREIGNOWNERSHIPINTHESECTORISNOWNEGATIVEANDSTATISTICALLYSIGNIFICANTTHEPOINTESTIMATEINDICATESTHATTHEPRODUCTIVITYOFDOMESTICPLANTSIS45PERCENTAGEPOINT
29、SLOWERINANINDUSTRYWITH10PERCENTAGEPOINTHIGHERFOREIGNSHARETHECOEFFICIENTONPLANT_DFIISSTILLPOSITIVEBUTSMALLERINMAGNITUDE,ANDTHECOEFFICIENTONTHEINTERACTIONOFPLANT_DFIANDSECTOR_DFIISNOWRENDEREDINSIGNIFICANTTHETHIRDCOLUMNINCLUDESTHEINDUSTRYDUMMIESBUTOMITSTHEFACTORINPUTSTHEREASONWEDOTHISISTOEXAMINETHEEFFE
30、CTOFFOREIGNOWNERSHIPONTHESCALEOFDOMESTICPLANTSINTHETHIRDCOLUMN,THECOEFFICIENTONTHEFOREIGNSHAREINTHEINDUSTRYISLARGEANDNEGATIVEANDPRECISELYESTIMATEDTHEPOINTESTIMATEINDICATESTHATA10PERCENTINCREASEINFOREIGNOWNERSHIPSHARELOWERSTHEOUTPUTOFDOMESTICFIRMSBY35PERCENTTHISSUGGESTSTHATFOREIGNOWNERSHIPFORCESDOMES
31、TICPLANTSTOCONTRACTTHELASTFOURCOLUMNSINTABLE1EXAMINETHEDYNAMICEFFECTOFFOREIGNINVESTMENTBYTAKINGDIFFERENCESOFTHEDATACOLUMN5TAKESFIRSTDIFFERENCES,COLUMN6TAKESSECONDDIFFERENCES,COLUMN7TAKESTHIRDDIFFERENCES,ANDCOLUMN8TAKESFOURDIFFERENCESOFTHEDATATHECOEFFICIENTONTHESHAREOFFOREIGNFIRMSINTHESECTORSECTOR_DF
32、IBECAMESMORENEGATIVEASWEMOVEFROMTHEFIRSTDIFFERENCESPECIFICATIONTOTHEFOURTHDIFFERENCESPECIFICATIONTHISSUGGESTSTHATTHENEGATIVEEFFECTOFFOREIGNOWNERSHIPONDOMESTICPRODUCTIVITYAPPEARSTOBECOMEMOREPRONOUNCEDOVERTIMETHECOEFFICIENTONPLANT_DFIREMAINSSMALLANDSTATISTICALLYINSIGNIFICANT,ASISTHECOEFFICIENTONTHEVAR
33、IABLEINTERACTINGPLANT_DFIANDSECTOR_DFI4ARETHEEFFECTSOFFOREIGNFIRMSLOCALIZEDWENOWEXPLOREWHETHERFOREIGNCAPITALONLYHASANEFFECTONTHEPRODUCTIVITYOFLOCALFIRMSINTHELOCALAREASOFAR,WEHAVESEARCHEDFORTHEEFFECTOFFOREIGNFIRMSONTHEPRODUCTIVITYOFDOMESTICFIRMSINTHESAMEINDUSTRYREGARDLESSOFTHEGEOGRAPHICLOCATIONOFTHEF
34、OREIGNFIRMOROFTHEDOMESTICFIRMHOWEVER,THETEXTILESECTORINSHANGHAIWILLHAVEMUCHOFANEFFECTONTHEPRODUCTIVITYOFADOMESTICFIRMINTHESAMETEXTILEINAFARAWAYPROVINCESUCHASGANSUTOEXAMINETHIS,WEENTERAVARIABLEFORTHESHAREOFFOREIGNFIRMSINAREGIONDEFINEDASACHINESEPROVINCEINEQUATION1THATIS,WEDEFINEAVARIABLELOCAL_SECTOR_D
35、FIASTHEEMPLOYMENTSHAREOFFOREIGNFIRMSINAGIVENCHINESEPROVINCETHENATURALISSUEWITHINTERPRETINGTHECOEFFICIENTONTHISVARIABLEASREFLECTINGTHEEFFECTOFLOCALSPILLOVERSOFFOREIGNFIRMSISTHATFOREIGNFIRMSCOULDCHOOSETOLOCATEINPROVINCESWITHCHARACTERISTICSTHATALSOLEADTOHIGHERDOMESTICPRODUCTIVITYTHATIS,THERECOULDBEANOM
36、ITTEDVARIABLE,SUCHASTHEQUALITYOFLOCALINFRASTRUCTURETHATAFFECTSBOTHTHEDECISIONOFFOREIGNFIRMSTOLOCATEINAGIVENPROVINCEANDTHEPRODUCTIVITYOFDOMESTICFIRMSWEDEALWITHTHISOMITTEDVARIABLEBIASINTWOWAYSFIRST,WECONTROLFORREGIONALGDPTHESEESTIMATESARESEENINTHESECONDCOLUMNINTABLE2ASCANBESEEN,THERESEEMSTOBECLEAREVID
37、ENCETHATFOREIGNPRESENCEINAREGIONHASALARGEANDPOSITIVEEFFECTONTHEPRODUCTIVITYOFDOMESTICFIRMSASECONDAPPROACHWETAKEISTOESTIMATE“WITHIN”ESTIMATESBYSUBTRACTINGEACHVARIABLEFROMITSPLANTSPECIFICMEANOVERTIMETHEIDEAISTHATTHISWOULDDEALTHEOMITTEDFIXEDCHARACTERISTICSOFEACHPLANTBYONLYUSINGTHEDEVIATIONOFTHEVARIABLE
38、SFROMITSMEANOVERTIMEHERE,THEESTIMATESPROVIDELITTLEEVIDENCETHATFOREIGNPENETRATIONHASAPOSITIVEEFFECTONTHEPRODUCTIVITYOFDOMESTICFIRMSINTHEPROVINCETHISDOESNOTCHANGEEVENWHENPROVINCIALGDPISINCLUDEDONTHERIGHTHANDSIDEOFTHEREGRESSION5DOTHEEFFECTOFFOREIGNFIRMSDIFFERBYPLANTSIZEWENOWPRESENTESTIMATESSEPARATELYFO
39、RLARGEANDSMALLPLANTSHERE,WEDEFINELARGEPLANTSASTHOSEWITHMORETHAN125EMPLOYEESASCANBESEEN,THEREISGENERALLYLITTLEDIFFERENCEINTHEOWNPLANTEFFECTBETWEENLARGEANDSMALLPLANTSFOREXAMPLE,THEOLSESTIMATEONPLANT_DFIISVIRTUALLYTHESAMEFORSMALLPLANTSASINLARGEPLANTSINBOTHCASES,THEPOINTESTIMATESUGGESTSTHATFOREIGNOWNEDP
40、LANTSAREGENERALLYMOREPRODUCTIVETHANDOMESTICPLANTS,BYROUGHLY14PERCENTAGEPOINTSHOWEVER,THEEFFECTOFFOREIGNINVESTMENTINASECTORDOESSEEMTOCLEARLYDIFFERBETWEENLARGEANDSMALLPLANTS,ATLEASTWHENONECONSIDERSTHEOLSESTIMATESFORSMALLPLANTS,FOREIGNPRESENCEINTHESECTORAPPEARSTOBEASSOCIATEDWITHLOWERPRODUCTIVITYFORLARG
41、EPLANTS,THECONVERSEISTRUESPECIFICALLY,FORSMALLDOMESTICPLANTS,A10PERCENTAGEPOINTINCREASEINTHEEMPLOYMENTSHAREOFFOREIGNFIRMSINTHESECTORISASSOCIATEDWITHA12PERCENTAGEPOINTDECLINEINPRODUCTIVITYFORLARGEDOMESTICPLANTS,ASIMILARINCREASEINTHESHAREOFFOREIGNFIRMSINTHESECTORISASSOCIATEDWITHA1PERCENTAGEPOINTINCREA
42、SEINPRODUCTIVITYTHISFINDINGISIMPORTANTINUNDERSTANDINGTHEPREVIOUSFINDINGTHATFOREIGNPRESENCEINASECTORDOESNOTAPPEARTOHAVEANEFFECTONDOMESTICFIRMPRODUCTIVITYWHATWEWEREMEASURINGBEFOREWASTHEAGGREGATEEFFECT,BUTITCLEARLYMASKEDIMPORTANTHETEROGENEITYINTHEIMPACTOFFOREIGNCAPITALBETWEENLARGEANDSMALLPLANTS6CONCLUS
43、IONINTHISPAPER,WEHAVEEXPLOITEDAUNIQUEPANELDATASETOFCHINESEFIRMSINTHEMANUFACTURINGSECTORTOGETATTHEPERENNIALQUESTIONOFTHEEFFECTOFFOREIGNDIRECTINVESTMENTONTHEPRODUCTIVITYOFDOMESTICFIRMSCLEARLY,THISISATOPICOFMAJORRELEVANCEFORPOLICYMANYCOUNTRIESHAVEVIEWEDTHEATTRACTIONOFFOREIGNCAPITALASCENTRALTOTHEIRDEVEL
44、OPMENTSTRATEGY,ANDCHINAHASCERTAINLYBEENAMAJORPLAYERINTHISTHISPAPERFOLLOWSTHEAPPROACHTAKENBYAITKENANDHARRISON1999TOEXAMINETHEEFFECTOFFOREIGNCAPITALINCHINAOURFINDINGSSUGGESTTHATFOREIGNFIRMSAREINDEEDMOREPRODUCTIVETHANDOMESTICFIRMSWHENWEEXAMINETHEAVERAGEPRODUCTIVITYOFDOMESTICPLANTSINTHESAMESECTOR,WEFIND
45、LITTLEEVIDENCETHATFOREIGNPRESENCEHADAPOSITIVEEFFECTHOWEVER,WHENWEUNPACKTHISEFFECT,WEFINDCLEAREVIDENCETHATFOREIGNPRESENCEINASECTORHADALARGEANDPOSITIVEEFFECTONLARGEDOMESTICPLANTS,BUTHADANEGATIVEEFFECTONSMALLDOMESTICPLANTS译文中国国内的企业受益于外国直接投资吗摘要本文采用中国制造业的面板数据来衡量外国公司对国内企业生产率的影响。我们发现在同一行业中外国公司要比国内企业的生产率高很多
46、。但外国公司对国内制造业的平均生产率没有任何积极或消极的影响。然而，我们找到的明确证据表明，这个面板数据掩盖了重要的企业异质性国内大公司大大受益于现存的外国直接投资，而国内小公司则没能从中受益。一、引言近几十年来，外国直接投资（FDI）已经在发展中国家急速增长。事实上，许多决策者认为，吸引外商直接投资是加快国家经济发展的一个关键。论证焦点是，外国公司采用的新技术将帮助提高国内企业的技术能力。这是可以通过多种渠道实现的。例如，外国公司的中国员工可以学习该外国公司的新技术并且将这些技术传播到自己创办的公司或移动到其他国内公司。它可能在外国公司通过刺激上游和下游之间的联系以帮助提升国内公司技术。由于
47、以上原因，许多国家都推出了更好的税务优惠，从税收假日提供公用设施到其他在利率上的优惠，没有一个国家做的比中国更好了，那里的政府已经提出了许多奖励措施，吸引外国投资，至今已创造出不凡的成绩。事实上，中国已成为世界上外商投资的最大直接受益者。然而，却没有证据表明外国公司对中国国内的企业有任何消极的影响。最近有一个重要论文，由艾特肯和哈里森（1990）著，他们使用先进的计量技术，利用来自于委内瑞拉的企业数据来测算出外国企业对国内企业生产率的影响程度。最终，艾特肯和哈里森（1999）发现，外国直接投资对国内企业的生产率并没有影响。本文将沿用艾特肯和哈里森的方法，采用独特的面板数据来研究中国公司。具体来
48、说，有两个问题我们要解决。首先，在中国的合资或外国独资子公司比国内有类似部门的企业拥有更高的生产率水平吗第二，现存的外国企业（无论是合资或独资外国公司）是否将技术“外溢”到了东道国的企业我们提出三个结论。首先，外国公司的确比国内同一行业的企业更有效率。其次，一个行业的外国公司对国内公司的平均生产率并没有产生影响，无论是积极或消极的。第三，一个行业的国外渗透占重要的不平衡的影响国内大型企业似乎是受益于现存的外国公司，而国内小企业似乎并未受益。本文提供了大量关于外国投资对国内企业生产率产生影响的资料。除了上面提到的文件，其他的还包括BLOMSTROM（1986年）墨西哥制造业和加拿大的GLOBER
49、MAN（1979年）。至于中国，我们没看到有任何文章使用了一级的公司资料来研究外国投资对中国国内企业生产率的影响。因此，本文的主要贡献是首先证明了一个国家的外国资本带来的资本，而这些外国资本绝大部分来自于世界范围内的外商投资。本文的结构如下。第2节描述的数据用于分析。第3节在审查外国投资与生产率关系的基础上，提出实证结论。第4节审查外溢是否是地区特性。第5节寻找证据支持，外国资本对大企业和小企业有不同的影响。第6节结论。二、数据本文所使用的数据是公司第一手的数据，是由中国国家统计局统计的每年一次的中国工业调查。这些数据公布在中国统计年鉴工业那一章。数据是从1998年到2004年。这项调查覆盖了所有国有和非国有的收入大于5亿元的工厂。这些工厂大概有20万至25万个。该数据还提供了工厂标识符，因此我们能够随着时间跟踪这些工厂。对于每一个工厂，还提供了该厂的行业，地理位置，所有权，资本存量的账面价值，投入成本，收入，就业与收入。还有部分信息我们使用如下数据。首先，我们以三位数水平来确定工厂所在的产业。数据中大约有180个3位数产业。第二，我们使用工厂的收入，就业，投入成本，和资本存量的账面价值来衡量工厂的生产率。使用资本存量的账面价值而没有使用市场价值可能会成为一个关注的问题。我们试图处理的方法就是，用不变增长率的假设和资本存量的账面价值来替代资本的市场价值。结论如下，不
