1、1本科毕业论文外文翻译外文题目TRADEOPENNESSANDECONOMICGROWTHACROSSCOUNTRYEMPIRICALINVESTIGATION出处JOURNALOFDEVELOPMENTECONOMICS作者HALITYANIKKAYA原文TRADEOPENNESSANDECONOMICGROWTHACROSSCOUNTRYEMPIRICALINVESTIGATIONHALITYANIKKAYAABSTRACTTHISPAPERDEMONSTRATESTHATTRADELIBERALIZATIONDOESNOTHAVEASIMPLEANDSTRAIGHTFORWARDRELA
2、TIONSHIPWITHGROWTHUSINGALARGENUMBEROFOPENNESSMEASURESFORACROSSSECTIONOFCOUNTRIESOVERTHELASTTHREEDECADESWEUSETWOGROUPSOFTRADEOPENNESSMEASURESTHEREGRESSIONRESULTSFORNUMEROUSTRADEINTENSITYRATIOSAREMOSTLYCONSISTENTWITHTHEEXISTINGLITERATUREHOWEVER,CONTRARYTOTHECONVENTIONALVIEWONTHEGROWTHEFFECTSOFTRADEBAR
3、RIERS,OURESTIMATIONRESULTSSHOWTHATTRADEBARRIERSAREPOSITIVELYAND,INMOSTSPECIFICATIONS,SIGNIFICANTLYASSOCIATEDWITHGROWTH,ESPECIALLYFORDEVELOPINGCOUNTRIESANDTHEYARECONSISTENTWITHTHEFINDINGSOFTHEORETICALGROWTHANDDEVELOPMENTLITERATURE1INTRODUCTIONDOOPENECONOMIESGROWFASTERTHANCLOSEDECONOMIESALMOSTALLEMPIR
4、ICALGROWTHSTUDIESHAVEPROVIDEDANAFFIRMATIVEANSWERTOTHISQUESTIONTHEREASONFORTHISSTRONGBIASINFAVOROFTRADELIBERALIZATIONISPARTLYBASEDONTHECONCLUSIONSOFAWIDERANGEOFEMPIRICALSTUDIES,WHICHCLAIMEDTHATOUTWARDORIENTEDECONOMIESCONSISTENTLYHAVEHIGHERGROWTHRATESTHANINWARDORIENTEDCOUNTRIESITISALSOPARTLYDUETOTHETR
5、AGICFAILURESOFIMPORTSUBSTITUTIONSTRATEGIES,ESPECIALLYINTHE1980SANDOVERSTATEDEXPECTATIONSFROMTRADELIBERALIZATION2ACCORDINGLY,THISSTUDYEXAMINESTHEGROWTHEFFECTSOFALARGENUMBEROFMEASURESOFTRADEOPENNESSUSINGTHESAMESPECIFICATIONFORVARIOUSOPENNESSMEASURESWHEREEACHVARIABLEMEASURESADIFFERENTASPECTOFOPENNESS,W
6、ETRYTOACCOMPLISHTWOTHINGSFIRST,WEBELIEVETHATOURRESULTSPROVIDEAMORECOMPLETEPICTUREOFTHERELATIONSHIPBETWEENOPENNESSANDGROWTHASCOMPAREDTOEMPIRICALSTUDIESTHATUSEDASUBSETOFOPENNESSMEASURESSECOND,ITENABLESUSTOCOMPAREREGRESSIONRESULTSACROSSINDIVIDUALMEASURESANDACROSSGROUPSWEDIVIDEOURTRADEOPENNESSMEASURESIN
7、TOTWOBROADCATEGORIESMEASURESOFTRADEVOLUMESANDMEASURESOFTRADERESTRICTIONSONTHEONEHAND,OURESTIMATIONRESULTSFORVARIOUSMEASURESOFTRADEVOLUMESINDICATETHATTHEREISAPOSITIVEANDSIGNIFICANTASSOCIATIONBETWEENTRADEOPENNESSANDGROWTHTHEYAREALSOCONSISTENTWITHTHECONCLUSIONSOFEMPIRICALANDTHEORETICALGROWTHSTUDIESONTH
8、EOTHERHAND,OURESTIMATIONRESULTSFORTRADEBARRIERSCONTRADICTTHECONVENTIONALVIEWONTHEGROWTHEFFECTSOFTRADERESTRICTIONS,WHICHSUGGESTSANADVERSEASSOCIATIONBETWEENTRADEBARRIERSANDGROWTHTHEOUTLINEOFTHISPAPERISASFOLLOWSTHENEXTSECTIONREVIEWSTHETHEORETICALANDEMPIRICALLITERATUREONTRADEOPENNESSSECTION3DESCRIBESAST
9、ANDARDGROWTHEQUATIONANDTHEDATASOURCESANDDEFINITIONSSECTION4SEPARATELYREPORTSTHEESTIMATIONRESULTSFORVARIOUSMEASURESOFTRADEFLOWSANDRESTRICTIONSONTHESEFLOWSINTHISSECTION,WEFURTHERADDRESSTHEPOTENTIALPROBLEMSOFTHECROSSCOUNTRYGROWTHFRAMEWORKSECTION5CONCLUDESTHEPAPER2LITERATUREREVIEWTHERELATIONSHIPBETWEENT
10、RADEOPENNESSANDGROWTHISAHIGHLYDEBATEDTOPICINTHEGROWTHANDDEVELOPMENTLITERATUREYET,THISISSUEISFARFROMBEINGRESOLVEDTHEORETICALGROWTHSTUDIESSUGGESTATBESTAVERYCOMPLEXANDAMBIGUOUSRELATIONSHIPBETWEENTRADERESTRICTIONSANDGROWTHTHEENDOGENOUSGROWTHLITERATUREHASBEENDIVERSEENOUGHTOPROVIDEADIFFERENTARRAYOFMODELSI
11、NWHICHTRADERESTRICTIONSCANDECREASEORINCREASETHEWORLDWIDERATEOFGROWTHSEEROMER,1990GROSSMANANDHELPMAN,1990RIVERABATIZANDROMER,1991A,BMATSUYAMA,1992NOTETHATIFTRADINGPARTNERSAREASYMMETRICCOUNTRIESINTHESENSETHATTHEYHAVECONSIDERABLYDIFFERENTTECHNOLOGIESANDENDOWMENTS,EVENIFECONOMICINTEGRATIONRAISESTHEWORLD
12、WIDEGROWTHRATE,ITMAYADVERSELYAFFECTINDIVIDUALCOUNTRIESSEEGROSSMANANDHELPMAN,1991A,BLUCAS,1988RIVERABATIZAND3XIE,1993YOUNG1991ITISWORTHWHILETONOTETHATTHETHEORETICALGROWTHLITERATUREHASGIVENMOREATTENTIONTOTHERELATIONSHIPBETWEENTRADEPOLICIESANDGROWTHRATHERTHANTHERELATIONSHIPBETWEENTRADEVOLUMESANDGROWTHT
13、HEREFORE,THECONCLUSIONABOUTTHERELATIONSHIPBETWEENTRADEBARRIERSANDGROWTHCANNOTBEDIRECTLYAPPLIEDTOTHEEFFECTSOFCHANGESINTRADEVOLUMESONGROWTHEVENTHOUGHTHESETWOCONCEPTS,TRADEVOLUMESANDTRADERESTRICTIONS,AREVERYCLOSELYRELATED,THEIRRELATIONSHIPWITHGROWTHMAYDIFFERCONSIDERABLYTHISISBECAUSETHEREARESEVERALOTHER
14、VERYIMPORTANTFACTORSTHATAFFECTACOUNTRYSEXTERNALSECTOR,SUCHASGEOGRAPHICALFACTORS,COUNTRYSIZE,ANDINCOMEINOTHERWORDS,ONESHOULDBEASCLEARASPOSSIBLEABOUTWHICHOPENNESSMEASUREHEUSESANDWHATARETHEEXACTMECHANISMSTHROUGHWHICHITAFFECTSTHEGROWTHWESHALLDISCUSSEACHOPENNESSMEASUREUSEDINTHISSTUDYLATERINTHESECTIONINTH
15、ETHEORYOFINTERNATIONALTRADE,THESTATICGAINSFROMTRADEANDLOSSESFROMTRADERESTRICTIONSHAVEBEENEXAMINEDTHOROUGHLYYET,TRADETHEORYPROVIDESLITTLEGUIDELINEASTOTHEEFFECTSOFINTERNATIONALTRADEONGROWTHANDTECHNICALPROGRESSONTHECONTRARY,THENEWTRADETHEORYMAKESITCLEARTHATTHEGAINSFROMTRADECANARISEFROMSEVERALFUNDAMENTA
16、LSOURCESDIFFERENCESINCOMPARATIVEADVANTAGEANDECONOMYWIDEINCREASINGRETURNS3MODELANDDATAWEUSETHEFOLLOWINGEMPIRICALFRAMEWORKTOINVESTIGATELONGRUNGROWTHINGENERALFORM,THISMODELCANBECHARACTERIZEDASYTFYT,KT,HTZT,1WHEREYTISACOUNTRYSPERCAPITAGROWTHRATEINPERIODT,YTISINITIALGDPPERCAPITA,KTISTHEPHYSICALCAPITALSTO
17、CKPERPERSON,HTISINITIALHUMANCAPITALPERPERSONWEUSETELEPHONEMAINLINESPERWORKERANDLIFEEXPECTANCYRATESASROUGHPROXIESFORTHESTOCKOFPHYSICALANDHUMANCAPITAL,RESPECTIVELYALTHOUGHTHEINITIALGDPPERCAPITALEVELISEMPLOYEDTOASSESSTHEISSUEOFCONDITIONALCONVERGENCE,ITISPOSSIBLETOINTERPRETITASAPROXYFORTHESTOCKOFCAPITAL
18、FORACOUNTRYTHEVARIABLEZREPRESENTSAVECTOROFCONTROLANDENVIRONMENTALVARIABLESTHATAREPRIMARILYDETERMINEDBYDECISIONSOFGOVERNMENTSORINDIVIDUALSTHESEVARIABLESINCLUDEALARGENUMBEROFTRADEOPENNESSMEASURES,WARDEATHSANDTYPEOFREGIMENOTETHATWEALSOUSETWOGEOGRAPHICALFACTORS,AVARIABLETHATMEASURESWHETHERACOUNTRYIS4INA
19、TROPICALCLIMATEANDAVARIABLETHATMEASURESWHETHERACOUNTRYHASACCESSTOINTERNATIONALWATERWAYS4EMPIRICALRESULTS41CORRELATIONACROSSOPENNESSMEASURESTHERESULTSSHOWTHATTHEREISGENERALLYASTATISTICALLYSIGNIFICANTCORRELATIONAMONGTHEOPENNESSMEASURESMOREIMPORTANTLY,ALLOFTHECORRELATIONCOEFFICIENTSHAVETHECORRECTSIGNSE
20、XCEPTFORTHECORRELATIONCOEFFICIENTSOFTAXTRDWITHOPENANDUSBTRD,WHICHHAVEINCORRECTSIGNSBUTAREINSIGNIFICANTALTHOUGHTHERELATIONSHIPSBETWEENOPENNESSMEASURESWITHINEACHGROUPAREGENERALLYSTATISTICALLYSIGNIFICANTWITHTHECORRECTSIGNS,THERELATIONSHIPSBETWEENTHEINDICATORSACROSSTHEGROUPSTENDTOBEWEAKTHUS,ITISPROBABLY
21、SAFETOCONCLUDETHATTRADEBARRIERSMAYHAVENEGATIVEREPERCUSSIONSONGROWTHTHROUGHREDUCINGTHESIZEOFTHEEXTERNALSECTOROFACOUNTRY42TRADEVOLUMESANDGROWTHCONSISTENTWITHANUMBEROFEMPIRICALSTUDIES,ASREVIEWEDINHARRISON1996,OURRESULTSSUPPORTTHEHYPOTHESISTHATCOUNTRIESWITHHIGHERTRADESHARESARELIKELYTOGROWFASTERTHANOTHER
22、COUNTRIESONEOFTHEIMPORTANTCRITICISMSOFTHESIGNIFICANTCOEFFICIENTISTHATITDOESNOTNECESSARILYMEANTHEDIRECTIONOFCAUSATIONGOESFROMHIGHERTRADESHARESTOHIGHERGROWTHRATESMANYSTUDIESHAVEEMPHASIZEDTHELIKELIHOODOFREVERSECAUSATIONBETWEENGROWTHANDOPENBECAUSEITISEASYTOTHINKOFACASEINWHICHFASTGROWTHMAYCAUSEHIGHERTRAD
23、ESHARESTOTESTWHETHERTHESIMULTANEITYISSTRONGENOUGHTODRIVETHISRELATIONSHIP,WEUSETHEAVERAGESOFFIVEPREVIOUSYEARSINSTEADOFCONTEMPORANEOUSVALUES43TRADERESTRICTIONSANDGROWTHITISCLEARFROMTHEPRECEDINGSECTIONANDALSOFROMEXISTINGEMPIRICALSTUDIESTHATTRADEVOLUMESAREPOSITIVELYANDSIGNIFICANTLYCORRELATEDWITHGROWTHCO
24、NSEQUENTLY,ONEWOULDEXPECTTHATANYTHINGTHATPOSESABARRIERTOINTERNATIONALTRADEISLIKELYTOBEHARMFULTOLONGRUNGROWTHINOTHERWORDS,BARRIERSTOTRADEINTHEFORMSOFTARIFFS,EXPORTDUTIESANDTAXESONINTERNATIONALTRADEAREEXPECTEDTOBECORRELATEDWITHGROWTHNEGATIVELYDUETOTHEIRPOTENTIALTRADEREDUCINGEFFECTSMOREOVER,ALARGENUMBE
25、ROFSTUDIESREPORTEDANEGATIVERELATIONSHIP5BETWEENGROWTHANDAVERAGETARIFFSANDALSOANUMBEROTHERFORMSOFTRADERESTRICTIONS44TRADEVOLUMESANDTRADEBARRIERSOURESTIMATIONRESULTSFORTRADEINTENSITYRATIOSANDTRADEBARRIERSMAYSEEMINCONSISTENTWITHONEANOTHERBECAUSETHEYSEEMTOPROVIDECOMPLETELYDIFFERENTANDCONTRADICTORYANSWER
26、STOTHEQUESTIONCONCERNINGTHEGROWTHEFFECTSOFTRADEOPENNESSTOSHEDSOMELIGHTSONTHESESEEMINGLYCONTRADICTORYRESULTS,WEPROCEEDINTWOWAYSFIRST,WEREESTIMATETHEREGRESSIONSFORTRADEBARRIERSINTABLE3USINGTRADESHARESOPENINSTEADOFPOPULATIONDENSITIESDENSITYANDREPORTTHERESULTSINTHEFIRSTCOLUMNOFTABLE4SECOND,WEREGRESSTRAD
27、ESHARESOPENONTHEMEASURESOFTRADEBARRIERSUSEDTHROUGHOUTTHESTUDYANDONOTHERVARIABLESSEETABLE4FORTHECONTROLVARIABLESTHUS,THESERESULTSSUGGESTTHATWHILETRADEBARRIERSHAVEADVERSEEFFECTSONGROWTHTHROUGHREDUCINGTRADE,THEYPOSITIVELYAFFECTGROWTHTHROUGHSUPERIORRESOURCEALLOCATIONAND/ORPOSITIVEEXTERNALITIESOURRESULTS
28、IMPLYTHATTHELATTERDOMINATESTHEFORMER45SENSITIVITYANALYSISINTHISSECTION,WECARRYOUTSEVERALROBUSTNESSTESTSFIRST,THEREGRESSIONSRESULTSPRESENTEDHEREMAYBESUBJECTTOSIMULTANEITYPROBLEMSTHUS,WETESTTHESENSITIVITYOFOURRESULTSFOROPENNESSMEASURESBYTAKINGINTOACCOUNTENDOGENEITYNEXT,WEALSOREPLICATEOURRESULTSINTABLE
29、5USING5YEARAVERAGESINSTEADOFDECADALAVERAGESANDREPORTTHEMINAPPENDIXBFINALLY,WEEMPLOYECONOMETRICMETHODSTHATCONTROLFORFIXEDFACTORSFIRSTDIFFERENCESANDFIXEDEFFECTS5CONCLUSIONTHISPAPERINVESTIGATESTHERELATIONSHIPBETWEENAWIDEVARIETYOFTRADEOPENNESSMEASURESANDGROWTHWEBASICALLYUSETWOTYPESOFOPENNESSMEASURESTHEF
30、IRSTGROUPISVARIOUSMEASURESOFTRADEVOLUMESEXCEPTPOPULATIONDENSITIESTRADESHARES,EXPORTSHARES,ANDIMPORTSHARESINGDPAREWIDELYUSEDINTHELITERATUREANDAREFOUNDTOBESIGNIFICANTLYANDPOSITIVELYCORRELATEDWITHGROWTHHOWEVER,MUCHLESSATTENTIONHASBEENGIVENTODISAGGREGATEDMEASURESOFTRADEINTENSITYINTHEEMPIRICALLITERATURET
31、HUS,THISSTUDYEMPLOYSSEVERALNEWMEASURESOFTRADEINTENSITY,WHICHARETRADEWITHOECDCOUNTRIES,NONOECDCOUNTRIES,ANDUSBILATERALTRADEFIGURESSIGNIFICANT,POSITIVECOEFFICIENTSFORTHESE6MEASURESSHOWTHATTHEGROWTHEFFECTSOFTRADEWITHDEVELOPEDCOUNTRIESARENOTCONSIDERABLYDIFFERENTFROMTRADEWITHDEVELOPINGCOUNTRIESFURTHER,PO
32、PULATIONDENSITIESALSOPOSITIVELYAFFECTGROWTHTHROUGHINCREASINGTRADEVOLUMESHENCE,THEREGRESSIONRESULTSFORTRADEVOLUMESPROVIDESUBSTANTIALSUPPORTFORTHEHYPOTHESISTHATTRADEPROMOTESGROWTHTHROUGHANUMBEROFCHANNELSSUCHASTECHNOLOGYTRANSFERS,SCALEECONOMIES,ANDCOMPARATIVEADVANTAGEANDARECONSISTENTWITHEARLIEREMPIRICA
33、LSTUDIESOFTHEISSUESURPRISINGLY,UNLIKETHELITERATUREONTHEGROWTHEFFECTSOFTRADEINTENSITYRATIOS,FINDINGSOFEMPIRICALSTUDIESOFTRADERESTRICTIONSARECONSIDERABLYDIFFERENTFROMPREDICTIONSOFTHEORETICALSTUDIESEVENTHOUGHTHETHEORETICALGROWTHSTUDIESPROVIDEDNOCONCLUSIVEEVIDENCEABOUTTHEDIRECTIONOFGROWTHEFFECTSOFTRADEB
34、ARRIERS,ESPECIALLYFORDEVELOPINGCOUNTRIES,AGREATMAJORITYOFTHEEMPIRICALSTUDIESCONCLUDEDTHATTHEREEXISTSASIGNIFICANTANDNEGATIVERELATIONSHIPBETWEENTRADERESTRICTIONSANDGROWTHWHEREAS,OURRESULTSAREMUCHCLOSERTOTHEPREDICTIONSOFTHEORETICALSTUDIESANDEVIDENTLYCONTRADICTTHEFINDINGSOFEARLIEREMPIRICALSTUDIESWEBELIE
35、VETHATOURRESULTSCASTSUBSTANTIALDOUBTSONTHECONVENTIONALVIEWTHATSUGGESTSAROBUSTANDNEGATIVERELATIONSHIPBETWEENTRADEBARRIERSANDGROWTHINOTHERWORDS,ALLMEASURESOFTRADEBARRIERSUSEDINTHESTUDYARESIGNIFICANTLYANDPOSITIVELYCORRELATEDWITHGROWTHEXCEPTFORRESTRICTIONSONCURRENTACCOUNTPAYMENTS,WHICHISNEGATIVELYBUTINS
36、IGNIFICANTLYCORRELATEDWITHGROWTHTHUS,OURRESULTSACTUALLYPROVIDECONSIDERABLEEVIDENCEFORTHEHYPOTHESISTHATRESTRICTIONSONTRADECANPROMOTEGROWTH,ESPECIALLYOFDEVELOPINGCOUNTRIESUNDERCERTAINCONDITIONSITISCRUCIALTONOTETHATINTHISSTUDY,WEHAVENOINTENTIONOFESTABLISHINGASIMPLEANDSTRAIGHTFORWARDPOSITIVEASSOCIATIONB
37、ETWEENBARRIERSTOTRADEANDGROWTHRATHER,OURMAINGOALISTOSHOWTHATTHEREISNOSUCHRELATIONSHIPBETWEENTRADERESTRICTIONSANDGROWTHONTHECONTRARY,THISRELATIONSHIPMOSTLYDEPENDSONCERTAINCHARACTERISTICSOFCOUNTRIESINOTHERWORDS,RESTRICTIONSONTRADECANBENEFITACOUNTRYDEPENDINGONWHETHERITISADEVELOPEDORDEVELOPINGCOUNTRY,WH
38、ETHERITISABIGORSMALLCOUNTRY,ANDWHETHERACOUNTRYHASCOMPARATIVEADVANTAGEINTHOSESECTORSTHATARERECEIVINGPROTECTIONFURTHERMORE,THISSTUDYHASALSOADDRESSEDPOTENTIALSTATISTICALPROBLEMSINHERENTTOTHISSORTOFESTIMATIONANDTOINTERPRETINGTHESERESULTSTHUS,WECARRYOUTANUMBEROFROBUSTNESSTESTSASDISCUSSEDINTHETEXTOURRESUL
39、TSDONOTSEEMTOSENSITIVETOTHEDIFFERENT7STATISTICALMETHODS,SPECIFICATIONS,DATASETS,ANDOUTLIERPROBLEMSALTHOUGHTHEOLS,SUR,AND3SLSESTIMATESSOMEWHATSIMILARTOEACHOTHER,THEFIXEDEFFECTSESTIMATESIMPLYAMUCHWEAKERRELATIONSHIPBETWEENGROWTHANDTRADEOPENNESS,ESPECIALLYFORTRADEBARRIERS译文贸易开放和经济增长一个跨国实证调查HALITYANIKKAY
40、A摘要本文将通过在过去三十年中跨地区的国家所运用的大量开放措施来说明贸易自由化与增长之间并不是简单的明确的关系。我们将使用两组贸易开放措施。许多贸易强度比的回归结果大多符合现有的文献。然而，与贸易壁垒对增长影响的传统观念不同，本文的估计结果将显示，在大多数情况下，特别是对发展中国家来说，贸易壁垒和增长是明确地，显著地联系在一起的，这和增长和发展的理论研究结果也是一致的。1、引言开放的贸易比封闭的贸易增长的更快吗几乎所有最增长的实证研究都对这个问题做出了肯定的回答。这种深刻的偏见的产生一部分是由广泛的实证研究得出的结论所造成的，这些实证研究声称与内向型的经济体相比，外向型的经济体持续的拥有更高的
41、增长率。另一部分原因则是因为进口替代策略的戏剧性失败，特别是进口替代策略盛行的80年代，和对贸易自由化的过高期望。因此，本文将研究大量的贸易开放政策对增长的影响。各种各样的开放措施中每一个变量都测量着开放的不同方面。我们将运用相同的规格来试图完成两件事。首先，和运用开放措施的子集的实证研究相比，我们的研究结果将提供一个关于开放和增长之间关系的更详细的描述。其次，我们将对个别的措施和分组的措施分别进行回归结果比较。我们将贸易开放措施分为两大类贸易量措施和贸易管制措施。一方面，针对不同的贸易量的估计结果显示贸易开放和增长之间有着显著的正相关联系，这个估计结果也和研究增长的实证和理论的研究结果相一致
42、。另一方面，传统观点认为贸易壁垒不利于增长，而关于贸易壁垒的估计结果与贸易限制对增长的影响作用的传统观点相8违背。本文的概要如下。第二部分将回顾关于贸易开放的理论和实证文献。第三部分描述了一个标准增长方程、数据源和以及它的定义。第四部分个别地提到了关于贸易往来和贸易往来限制的各种措施的估计结果。在这部分里，我们将深入探讨跨国增长框架的潜在问题。第五部分是对本文的总结。2文献回顾在有关贸易和发展的文献中，贸易开放和增长之间的关系是其中频繁讨论的话题之一。但是这个问题还远远没有解决。有关增长的理论研究提出贸易管制和增长之间的关系充其量是复杂的且模糊不清的。研究内生增长的文献十分多样化，足以提供一系
43、列的模型，这些模型中，贸易管制可以增加或者减少全球的增长率（见罗默1990，格鲁斯曼和普曼1990，里维拉和罗默1991，松山1992）。如果就拥有大量不同技术和资源禀赋的层面上来说，两个贸易伙伴国家是不对称的，即使经济一体化提高了全球增长率，它还是可能对个别国家造成不利的影响。值得注意的是，研究增长的理论文献将研究重点放在了贸易政策和增长的关系上，而不是贸易额和增长的关系。因此关于贸易壁垒和增长关系的结论不能直接应用于贸易额的变化对增长的影响。尽管贸易量和贸易管制这两个概念联系的十分紧密，但他们各自和增长之间的关系可能十分不同。这是因为影响一国的对外部门的还有其他一些重要的因素，例如地理因素
44、，国家大小，以及收入等。换句话说，一个国家应该尽可能清楚他所使用的开放措施以及这种措施是通过什么确切的机制来影响增长的。在这部分我们讨论本文中所用到的各个开放措施。研究国际贸易的理论中，已经彻底地研究了贸易的静态收益和贸易管制的损失。但是这些贸易理论对于国际贸易对增长和技术进步的影响提到的却很少。相反地，新贸易理论明确了贸易收益有几个基本的来源比较优势的差异和经济收益递增。3模型和数据本文将运用下面的经验框架来研究长期增长。模型的一般形式如下YTFYT,KT,HTZT,1其中YT表示在一段时间内一国的人均增长率，YT表示初始的人均国内生产总值，KT表示人均物质资本存量，HT表示初始的人均人力资
45、本。我们把每个工人的电话和预期寿命率分别粗略的代表物质资本存量和人力资本存量。尽管初始的人均国内生产总值是9用来评估条件趋同这个问题的，但是对于一个国家来说，它也可能代表资本存量。变量Z代表了一个控制向量和环境变量，其中决定环境变量的首要因素是政府或个人的决定。环境变量包括大量的贸易开放措施，战争伤亡和政权类型。4实证结果41开放措施间的相关性结果显示，在统计意义上开放措施之间上通常是显著相关的。更重要的是，除了少数不正确的但是不重要的相关系数外，几乎所有的相关系数看上去都是正确的。尽管从统计意义上来说开放措施和每个变量之间的关系通常是显著正确的，但是每个变量的指标之间的关系却并不显著。因此，
46、只能得出这样的结论，即贸易壁垒可能通过缩减一个国家的对外部门规模来对增长产生消极的影响。42贸易量和增长和许多实证研究相一致，如哈里森（1996）文中所回顾的，本文的结果支持有更高贸易份额的国家比其他的国家增长地更快这个假设。对于显著变量的主要批评之一是它并不一定意味着因果关系的走向是从更高的贸易份额到更高的增长率的。由于快速的增长可能会导致更多的贸易份额这种情况很容易让你理解，因此许多研究都强调了增长和开放之间逆向思维的可能性。为了测试这种同步性是否足以推动这种关系，本文将运用过去五年的平均价值，而不是同时期的价值。43贸易管制和增长从上一部分以及现有的实证研究可以清楚的看到贸易量和增长之间
47、呈现显著正相关的关系。因此可见任何会对国际贸易带来障碍的因素都将对长期增长产生危害。换句话说，通过对国际贸易加征进口税，出口税和关税的形式产生的贸易壁垒由于潜在的贸易递减效应将对增长产生消极影响。而且许多的文献验证了增长和平均关税以及其他贸易管制形式之间的负相关关系。44贸易量和贸易壁垒本文关于贸易强度比和贸易壁垒的估计结果可能看上去和另一个不一致，因为他们似乎对于贸易开放对于增长的影响这个问题提出了完全不一样甚至是相反的答案。本文将从两方面来说明这些看上去相反的结论。首先，我们运用了贸易量而不是人口密度来重新估计贸易壁垒的回归分析。其次，我们运用了在研究中使用的贸易壁垒措施和其他一些变量来对
48、贸易份额进行回归分析。因此，这些结果显示了尽管贸易壁垒10通过减少贸易对增长产生了负效应，但是却通过卓越的资源配置或/和正外部性对增长产生了积极的影响。我们的结果显示是后者决定了前者。45敏感性分析在这一部分，我们将进行几个稳健性测试。首先，这里所呈现的回归结果会受到同时性问题的影响。因此我们在测试回归结果对开放措施的敏感性时会将内生性因素考虑进去。其次，我们使用的是五年的平均数据而不是十年平均。最后，我们使用计量经济学的方法，控制固定的几个因素，一阶差分和固定的影响。5总结本文对各种各样的贸易开放措施和增长之间的关系进行了调查。在调查中主要使用了两种类型的开放措施。第一组是各种各样的贸易量措
49、施（除了人口密度）。国内生产总值中的贸易份额，出口份额和进口份额被广泛的用于文献并且被认为和增长显著地正相关。但是，实证文献中模块化的贸易密集度措施却很少受到关注。因此，本文将运用一些新的贸易密集度措施，这些措施是和经合组织的国家，非经合组织的国家以及美国的双边贸易数据所进行的交易。值得注意的是，这些措施中的正系数显示发达国家中贸易对增长的影响与发展中国家中并无明显不同。进一步说，人口密度也通过增加贸易量对增长产生了积极的作用。因此，贸易量的回归结果给贸易通过诸多渠道，比如技术转移，规模经济和比较优势等提升了增长这个假设提供了实质性的支持，并且和早期的有关这个问题的实证研究的结论也相一致。令人意外的是，和研究贸易强度比对增长影响的文献不同，用实证方法研究贸易管制得出的结论和理论研究中的预测大不相同。尽管研究增长的理论文献没有提供结论性的证据来证明贸易壁垒对增长的影响方向，尤其对发展中国家来说，很大一部分的实证研究总结道贸易管制和增长之间有着显著的负相关关系。然而，本文的结果更接近于理论研究的预测，并且明显地和早期的实证研究结果相反。本文的结果对贸易壁垒和增长之间稳定的负相关关系提出了实质性的怀疑。换句话说，除了经常账户支付限制显著地与增长负相关外，文中所用所有的贸易壁垒的措施都和增长有着显著的正相关联系。因此，本文的结果实际上提供了大量的证据来证明贸易管制能推动增长这个
