产业结构和经济稳定【外文翻译】.doc

1、1本科毕业论文外文翻译外文题目INDUSTRIALSTRUCTUREANDECONOMICSTABILITY出处APPLIEDECONOMICLETTERS作者SHERRILLSHAFFER原文MOTIVATEDBYPRIORRESULTSPREDICTINGCONTRASTINGLINKAGESBETWEENINDUSTRIALSTRUCTUREANDECONOMICSTABILITY,WEPRESENTEXPLORATORYEMPIRICALEVIDENCEONTHISIMPORTANTISSUECONSISTENTWITHTHETURNOVERHYPOTHESIS,WEFINDTHATE

2、MPLOYMENTGREWMORESTEADILYWHEREBUSINESSESTABLISHMENTSINALLSECTORSWERELARGER,SUGGESTINGANOFFSETTINGBENEFITTOTHEFIRSTMOMENTCOSTSOFESTABLISHMENTSIZEIDENTIFIEDBYPREVIOUSRESEARCHCONSISTENTWITHTHEJOBMATCHINGHYPOTHESIS,WEFINDTHATEMPLOYMENTGREWMORESTEADILYWHEREMOREESTABLISHMENTSPERCAPITAOPERATEDINALLSECTORSS

3、IMILARBUTLESSCONSISTENTRESULTSWEREALSOFOUNDREGARDINGTHESTABILITYOFINCOMEGROWTHIINTRODUCTIONANDBACKGROUNDTHEFUNDAMENTALIMPORTANCEOFECONOMICPERFORMANCEHASSPAWNEDANEXTENSIVELITERATUREONTHEEMPIRICALDETERMINANTSOFECONOMICGROWTHSEERAJANANDZINGALES,1998LEVINEETAL,2000FORPROMINENTEXAMPLESONERECENTDISCOVERYI

4、STHATCOMMUNITIESORLOCALREGIONSTENDTOEXPERIENCEMORERAPIDGROWTHOFINCOMEOREMPLOYMENTWHEREBUSINESSESARESMALLERSHAFFER,2002,2006A,BASEPARATESTRANDOFRESEARCH,MEANWHILE,HASDEMONSTRATEDTHEIMPORTANCEOFSTUDYINGTHEVOLATILITYOFGROWTHRATESRATHERTHANMERELYTHEIRMEANSRAMEYANDRAMEY,1995KURZ,2004ISMIHAN,2005BEKAERTET

5、AL,2006ITISRELEVANTINTHISREGARDTHATSEVERALSTUDIESDAVISANDHALTIWANGER,1992ROB,1995DAVISETAL,1996HAVEFOUNDTHATJOBSATSMALLERFIRMSTENDTOBELESSPERMANENTTHANATLARGERFIRMSWHILEILMAKUNNASETAL2005HAVESUGGESTEDTHATSUCHTURNOVERMAYACTUALLYBEONEREASONFORFASTERPRODUCTIVITYGROWTH,ANUNEXPLOREDIMPLICATIONISTHATTHEVO

6、LATILITYOFEMPLOYMENT2ANDINCOMEMAYBEHIGHERWHEREBUSINESSESARESMALLERTHETURNOVERHYPOTHESISCONVERSELY,HETEROGENEITYAMONGDECISIONMAKERSDUETOHUMANFALLIBILITYCANRESULTINGREATERVARIABILITYOFECONOMICPERFORMANCEINMORECENTRALIZEDDECISIONPROCESSESSAH,1991SAHANDSTIGLITZ,1991ALMEIDAANDFERREIRA,2002,POSSIBLYSUGGES

7、TINGTHATALOCALECONOMYMAYEXHIBITGREATERSTABILITYWHERETHEAVERAGEFIRMSARESMALLERORMORENUMEROUSASIMILAROUTCOMEISPREDICTEDTOTHEEXTENTTHATACENTRALIZEDDECISIONPROCESSESARELESSSUCCESSFULATMANAGINGCONFLICTANDBDISTRIBUTIONALCONFLICTSIMPAIREFFICIENTADJUSTMENTSTOEXOGENOUSSHOCKSRODRIK,1999ALMEIDAANDFERREIRA,2002

8、INVIEWOFTHESECONTRASTINGCONSIDERATIONS,ANIMPORTANTEMPIRICALQUESTIONISWHETHERESTABLISHMENTSIZEANDOTHERMEASURESOFINDUSTRIALSTRUCTUREMAYBESYSTEMATICALLYASSOCIATEDWITHSECONDMOMENTMEASURESOFECONOMICPERFORMANCETHISARTICLEACCORDINGLYPRESENTSPRELIMINARYEVIDENCEOFLINKAGESBETWEENSELECTEDMEASURESOFINDUSTRIALST

9、RUCTUREANDTHEVOLATILITYOFLOCALINCOMEANDEMPLOYMENTGROWTHRATESWEUSETHREEMEASURESOFINDUSTRIALSTRUCTURE,EACHDISTINGUISHEDBYBROADSECTORASINSHAFFER2002,WEMEASUREAVERAGEESTABLISHMENTSIZEBYNUMBEROFEMPLOYEESAND,ALTERNATIVELY,BYDOLLARSOFVALUEADDED,SHIPMENTSORRECEIPTSWEALSOLOOKATESTABLISHMENTSPERCAPITA,MOTIVAT

10、EDBYTWOOPPOSINGCONSIDERATIONSFINDINGANEWJOBSHOULDBEEASIERINAMARKETWITHMOREEMPLOYERSINAGIVENSECTOR,LEADINGTOMORESTABLELEVELSORGROWTHRATESOFINCOMEANDEMPLOYMENTTHEJOBMATCHINGHYPOTHESISBUT,CETERISPARIBUS,SMALLERFIRMSWILLTENDTOPERMITTHECOEXISTENCEOFLARGERNUMBERSOFFIRMS,INWHICHCASETHEDOCUMENTEDEMPLOYMENTT

11、URNOVERATSMALLERFIRMSDISCUSSEDABOVEWILLTENDTOOFFSETTHESTABILIZINGBENEFITOFMORENUMEROUSFIRMSEACHOFOURMEASURESOFINDUSTRIALSTRUCTUREISCOMPILEDSEPARATELYFORTHEMANUFACTURING,WHOLESALE,RETAILANDSERVICESECTORSWERELATETHESEMEASURESOFSTRUCTURETOTWOSECONDMOMENTMEASURESOFECONOMICPERFORMANCE,THESDOFANNUALREALPE

12、RCAPITAINCOMEGROWTHRATESANDTHESDOFTHEANNUALGROWTHRATEOFTOTALESTABLISHMENTEMPLOYMENTTHERESULTSCONTRIBUTETOTWOSEPARATESTRANDSOFTHELITERATURE,ONEMPIRICALCOVARIATESWITHGROWTHVOLATILITYANDONMACROECONOMICEFFECTSOFESTABLISHMENTSIZEANDOTHER3MEASURESOFINDUSTRIALSTRUCTUREWEFINDFORALLSECTORSTHATLARGERESTABLISH

13、MENTSANDMOREESTABLISHMENTSPERCAPITAAREASSOCIATEDWITHMORESTABLEEMPLOYMENTGROWTHRATES,CONSISTENTWITHTHETURNOVERANDJOBMATCHINGHYPOTHESESTHESAMELINKAGESAREFOUNDFORSOMEBUTNOTALLOFTHESECTORSWITHREGARDTOTHEVOLATILITYOFGROWTHRATESINREALPERCAPITAINCOMETHENEXTSECTIONINTRODUCESTHEEMPIRICALMODELANDTHESAMPLESECT

14、IONIIIPRESENTSTHERESULTS,WHILESECTIONIVCONCLUDESIITHEMODELANDSAMPLEWEEMBEDOURKEYVARIABLESINASTANDARDLINEAREMPIRICALGROWTHEQUATION,XSY1WHEREYISAMEASUREOFECONOMICPERFORMANCEASDISCUSSEDABOVE,ISANESTIMATEDINTERCEPTTERM,SISAMEASUREOFINDUSTRIALSTRUCTUREASDISCUSSEDABOVE,XISAVECTOROFCONTROLVARIABLESDISCUSSE

15、DBELOW，ANDAREESTIMATEDCOEFFICIENTSANDISASTOCHASTICERRORTERMASINBEKAERTETAL2006,OURSDSOFECONOMICGROWTHRATESAREMEASUREDOVERA5YEARPERIODASINPRIORSTUDIESOFECONOMICGROWTH,THECONTROLVECTORINCLUDESTHENATURALLOGARITHMOFPOPULATION,THEDENSITYOFPOPULATIONPERSQUAREMILEOFLANDAREA,AMEASUREOFEDUCATIONANDINITIALMED

16、IANHOUSEHOLDINCOMEPOPULATIONISAMEASUREOFMARKETSIZEASINCETORELLIANDGAMBERA2001ITISALSOSIMILARTOTHETOTALLABORFORCEVARIABLEUSEDINOHUALLACHAINANDSATTERTHWAITE1992ANDCANBEINTERPRETEDASMEASURINGURBANIZATIONECONOMIESIFJOBMATCHINGOCCURSMOREQUICKLYOREFFICIENTLYINMOREPOPULOUSAREAS,THENTHEESTIMATEDCOEFFICIENTO

17、NTHISVARIABLESHOULDBENEGATIVEANOTHERIMPLICATIONOFTHEJOBMATCHINGHYPOTHESISINTRODUCEDABOVEPOPULATIONDENSITYHASBEENFOUNDSIGNIFICANTLYRELATEDTOSEVERALFIRSTMOMENTMEASURESOFECONOMICPERFORMANCE,POSSIBLYDUETOSCALEEFFECTSORTOSUPERIORMATCHINGBETWEENFIRMSANDWORKERSINDENSERMARKETSCICCONEANDHALL,1996ANDERSSONETA

18、L,2004CARLINOETAL,2007STRUMSKYETAL,2005IFTHESEBENEFITSINFLUENCEECONOMICSTABILITY,ASONEMIGHTEXPECT,THENTHEESTIMATEDCOEFFICIENTON4POPULATIONDENSITYSHOULDBENEGATIVEINOURMODELEDUCATIONISMEASUREDASTHEPERCENTAGEOFPOPULATIONAGED25ANDOVERWHOHAVECOMPLETEDHIGHSCHOOL,ANDREFLECTSTHEACCUMULATEDLEVELOFHUMANCAPITA

19、LITISSIMILARTOMEASURESUSEDINPREVIOUSSTUDIESOFECONOMICGROWTHSUCHASRAJANANDZINGALES1998,LEVINEETAL2000ANDCETORELLIANDGAMBERA2001,WITHTHEORETICALLINKAGESTOAVERAGEGROWTHRATESEXPLOREDBYTELES2005ARELATEDMEASUREOFEDUCATIONHASBEENUSEDINATLEASTONESTUDYOFGROWTHVOLATILITYBEKAERTETAL,2006INADDITION,EDUCATIONWAS

20、FOUNDTOBEPOSITIVELYASSOCIATEDWITHSECTORALEMPLOYMENTGROWTHINUSMETROPOLITANAREASBYOHUALLACHAINANDSATTERTHWAITE1992IFEDUCATIONCONTRIBUTESTOECONOMICSTABILITYASWELL,ITSESTIMATEDCOEFFICIENTINOURREGRESSIONSSHOULDBENEGATIVEINITIALMEDIANHOUSEHOLDINCOMECANREFLECTACONVERGENCEEFFECTINFIRSTMOMENTMEASURESOFECONOM

21、ICPERFORMANCE,ASNOTEDBYBARROANDSALAIMARTIN1992THESAMELOGICWOULDNOTAPPLYTOSECONDMOMENTMEASURESOFECONOMICPERFORMANCE,RENDERINGTHESIGNANDSIGNIFICANCEOFTHEESTIMATEDCOEFFICIENTONTHISVARIABLEANOPENEMPIRICALQUESTIONHOWEVER,ASIMILARVARIABLEINITIALPERCAPITAGDPHASBEENUSEDINATLEASTONEPRIORSTUDYOFGROWTHVOLATILI

22、TYBEKAERTETAL,2006TABLE1SUMMARIZESTHEDATAOURSAMPLECOMPRISESMORETHAN2000USNONMETROPOLITANCOUNTIES,MEASURINGECONOMICPERFORMANCEDURING19911995ANDSTRUCTUREMEASURESASOF1987METROPOLITANAREASAREEXCLUDEDBECAUSETHEIRBORDERSAREGENERALLYNOTCOTERMINOUSWITHINDIVIDUALCOUNTIES,CONFOUNDINGMEASUREMENTPROBLEMSFORVARI

23、ABLESDRAWNFROMCOUNTYLEVELDATATHOUGHNOTREPORTEDINTHETABLE,THEPAIRWISECORRELATIONCOEFFICIENTSBETWEENAVERAGEESTABLISHMENTSIZEANDESTABLISHMENTSPERCAPITARANGEDBETWEEN021AND018,ANDWEREJUST007INTHEWHOLESALESECTORAND007INTHEMANUFACTURINGSECTORTHESESMALLANDHIGHLYVARIABLECORRELATIONSINDICATETHATTHETWOCATEGORI

24、ESOFSTRUCTUREVARIABLESREFLECTSTATISTICALLYSEPARATEDIMENSIONSOFINDUSTRIALSTRUCTUREINOURSAMPLETHESELECTIONOFPERFORMANCEDATAASOFSEVERALYEARSFOLLOWINGTHESTRUCTUREDATAHELPSTOREDUCETHELIKELIHOODOFREVERSECAUSALITYALTHOUGH,COMMONTOALLEMPIRICALGROWTHSTUDIES,CAUSALITYCANNOTBEDEFINITIVELYESTABLISHEDTHISLAGSTRU

25、CTUREALSOMINIMIZESTHEPOTENTIALFORENDOGENEITYBIAS,ASTHEREGRESSORSAREPREDETERMINED5IIIRESULTSTABLE2REPORTSTHEREGRESSIONESTIMATESFORTHESDOFREALPERCAPITAINCOMEGROWTHRATES,WHILETABLE3REPORTSESTIMATESFORTHESDOFEMPLOYMENTGROWTHRATESASINDUSTRIALSTRUCTUREISMEASUREDINTHREEWAYSFOREACHOFFOURSECTORS,EACHTABLEREP

26、ORTS12REGRESSIONSDUETOMISSINGORZEROESTABLISHMENTDATAFORAFEWCOUNTIESINEACHSECTOR,THEVARIOUSREGRESSIONSUTILIZESLIGHTLYDIFFERINGNUMBERSOFOBSERVATIONS,ASREPORTEDINTHETABLESTHESIGNIFICANCELEVELSARECOMPUTEDFROMSESCORRECTEDFORHETEROSCEDASTICITYINTABLE2,THEVARIOUSSTRUCTUREMEASURESARESTATISTICALLYSIGNIFICANT

27、INEIGHTOFTHE12REGRESSIONSPERCAPITAINCOMEISFOUNDTOGROWMORESTEADILYWHEREMANUFACTURING,RETAILANDWHOLESALEESTABLISHMENTSEMPLOYMOREWORKERS,ORWHEREWHOLESALEESTABLISHMENTSARELARGERASMEASUREDBYVALUEOFANNUALSHIPMENTSTHESEFINDINGSARECONSISTENTWITHTHETURNOVERHYPOTHESISDISCUSSEDABOVECONVERSELY,STEADIERGROWTHISS

28、EENWHEREWHOLESALEESTABLISHMENTSEMPLOYFEWERWORKERSORINCOUNTIESWITHFEWERWHOLESALEESTABLISHMENTSPERCAPITAORMOREMANUFACTURINGESTABLISHMENTSPERCAPITATHECONTRASTINGESTIMATESFORTHEWHOLESALESECTORMAYREFLECTNONLINEARITIESINTHEUNDERLYINGRELATIONSHIPS,ANISSUEBEYONDTHESCOPEOFTHISSTUDYTHESEFINDINGSINDICATETHATIN

29、DUSTRIALSTRUCTUREISSYSTEMATICALLYRELATEDTOTHEVOLATILITYOFINCOMEGROWTHRATES,INWAYSTHATARELARGELYCONSISTENTBUTTHATDOVARYSOMEWHATACROSSMAJORSECTORSTHEFITOFTHEEQUATIONSINTABLE2ISMODEST,WITHADJUSTEDR2RANGINGFROM03TONEARLY05THECONTROLVARIABLESAREMOSTLYSIGNIFICANTATTHE001LEVELINITIALMEDIANHOUSEHOLDINCOMEIS

30、NEVERSIGNIFICANT,INCONTRASTTOTHESIGNIFICANCEOFINITIALPERCAPITAGDPFOUNDINSEVERALCROSSCOUNTRYREGRESSIONSONCONSUMPTIONGROWTHVOLATILITYBYBEKAERTETAL2006INCOMEISFOUNDTOGROWMORESTEADILYINMOREPOPULOUSCOUNTIESPOPULATIONDENSITYANDEDUCATIONAREFOUNDTOCONTRIBUTETOHIGHERVOLATILITYOFPERCAPITAINCOMEGROWTHRATES,THO

31、UGHEDUCATIONLOSESITSSIGNIFICANCEWHENCONTROLLINGFORWHOLESALEESTABLISHMENTSPERCAPITATHESIGNOFTHEEDUCATIONVARIABLECONTRASTSWITHTHATFOUNDINCROSSCOUNTRYSPECIFICATIONSBYBEKAERTET6AL2006INTABLE3,THEVARIOUSSTRUCTUREMEASURESAREAGAINSTATISTICALLYSIGNIFICANTINEIGHTOFTHE12REGRESSIONSEMPLOYMENTISSEENTOGROWMOREST

32、EADILYWHERETHEREAREMOREESTABLISHMENTSPERCAPITAINANYOFTHEFOURSECTORS,CONSISTENTWITHTHEJOBMATCHINGHYPOTHESISINADDITION,STEADIEREMPLOYMENTGROWTHOCCURREDWHEREWHOLESALEORSERVICEESTABLISHMENTSEMPLOYMOREWORKERS,ORWHERERETAILORSERVICEESTABLISHMENTSHAVEALARGERAVERAGEDOLLARVOLUMEOFBUSINESS,CONSISTENTWITHTHETU

33、RNOVERHYPOTHESISTHESELATTERFINDINGSINDICATETHATTHEBENEFITSOFJOBLONGEVITYATLARGERFIRMSMORETHANOFFSETTHEVOLATILITYTHEORETICALLYPREDICTEDBYMORECENTRALIZEDDECISIONMAKING,ASDISCUSSEDABOVETHEFITOFTHESEEQUATIONSISLESSTIGHTTHANINTABLE2,WITHADJUSTEDR2AROUND01POPULATIONDENSITYISNEVERSIGNIFICANTBUTTHEOTHERCONT

34、ROLVARIABLESARECONSISTENTLYSIGNIFICANTEMPLOYMENTGROWSMORESTEADILYINMOREPOPULOUSCOUNTIES,WHEREEDUCATIONISHIGHER,ORWHEREINITIALMEDIANHOUSEHOLDINCOMEISLOWERTHEPOPULATIONRESULTISCONSISTENTWITHTHEJOBMATCHINGHYPOTHESISDISCUSSEDABOVE,ANDTHEEDUCATIONANDINITIALINCOMERESULTSMATCHTHOSEFOUNDINCROSSCOUNTRYREGRES

35、SIONSFORCONSUMPTIONGROWTHVOLATILITYBYBEKAERTETAL2006ASINTABLE2,EDUCATIONLOSESITSSIGNIFICANCEWHENCONTROLLINGFORWHOLESALEESTABLISHMENTSPERCAPITAIVCONCLUSIONMOTIVATEDBYPRIORRESULTSSUGGESTINGCONTRASTINGANDIMPORTANTPREDICTIONSRELATINGINDUSTRIALSTRUCTURETOECONOMICSTABILITY,WEHAVEEXPLOREDEMPIRICALLINKAGESB

36、ETWEENMULTIPLESECTORSPECIFICSTRUCTUREMEASURESANDTHEVOLATILITYOFSUBSEQUENTGROWTHRATESINREALPERCAPITAINCOMEANDEMPLOYMENTSIGNIFICANTASSOCIATIONSWEREFOUNDINTWOTHIRDSOFTHESPECIFICATIONS,CONTROLLINGFORASTANDARDSETOFENVIRONMENTALVARIABLESOVERALL,THETURNOVERHYPOTHESISWASFOUNDTODOMINATETHECONTRASTINGPREDICTI

37、ONSOFINEFFICIENTCENTRALIZEDDECISIONMAKING,ASEMPLOYMENTANDINCOMEBOTHTENDEDTOGROWMORESTEADILYWHEREESTABLISHMENTSWERELARGERTHESEFINDINGSSUGGESTANOFFSETTINGBENEFITTOTHEFIRSTMOMENTCOSTSOF7ESTABLISHMENTSIZEIDENTIFIEDBYSHAFFER2002,2006A,BTHEJOBMATCHINGHYPOTHESISISCONSISTENTWITHSTEADIERGROWTHWHEREMOREESTABL

38、ISHMENTSPERCAPITAOPERATE,ASFOUNDFORALLSECTORSINTHEEMPLOYMENTREGRESSIONSANDFORTHEMANUFACTURINGSECTORINTHEINCOMEREGRESSIONSCONTRASTINGRESULTSWEREFOUNDINTHEINCOMEREGRESSIONSFORWHOLESALEESTABLISHMENTSMEASUREDBYEMPLOYMENTSIZEANDFORWHOLESALEESTABLISHMENTSPERCAPITAALTHOUGHOURMULTIYEARLAGSTRUCTUREREDUCESTHE

39、LIKELIHOODOFREVERSECAUSALITYBETWEENINDUSTRIALSTRUCTUREANDGROWTHVOLATILITY,CAUSALITYCANNOTBEDEFINITIVELYESTABLISHED,ASHORTCOMINGCOMMONTOALLEMPIRICALGROWTHSTUDIESCONSEQUENTLY,POLICYINFERENCESSHOULDBEDRAWNWITHCAUTION,PENDINGADDITIONALRESEARCHONTHEPRECISEMECHANISMSUNDERLYINGTHEOBSERVEDPATTERNS译文产业结构和经济稳

40、定谢里尔谢弗经济及金融系电子邮箱SHAFFERUWYOEDU通过预测产业结构和经济稳定之间的联系，在事先对比结果的启发下，我们提出探索这一重要问题的证据。在符合营业额假设的前提下，我们发现更稳定的就业增长在各行业机构均较大，这意味着一个抵消造福于编制第一时刻由以前的研究中发现的成本。这与工作匹配假说相一致，即就业增长在经营各行业的场所更加稳定。与此相类似，但不太一致的结果也发现关于收入增长的稳定性。一、引言和背景经济运行的重要性催生了大量关于影响经济增长的实证因素的文献（见拉詹和辛格勒，1998；莱文等人，突出的例子为2000）。最近的一项研究发现，社区或局部地区往往经历着更快速的收入或就业增

41、长，尤其是在小规模企业（谢弗，2002，2006A，B）。一个单独的研究链，它同时展示了学习增长率波动的重要性，8而不仅仅是他们的手段（拉米和拉米，1995；库尔兹，2004；ISMIHAN，2005；贝卡尔特等，2006）。正是在这方面的几项有关的研究（戴维斯和黑尔蒂旺热，1992；罗布，1995；戴维斯等人，1996）发现，在大公司工作往往比在小企业更持久；而ILMAKUNNAS等（2005）建议，这类营业额实际上可能是一个更快的生产率增长的原因，一个未开发的含义是就业和收入的波动性可能会导致更高的营业额在规模较小的地方企业中（即营业额假说）。相反，决策者由于人类的不可靠而导致经济的差异性

42、更加集中地表现在决策的过程中（萨，1991；SAH和斯蒂格利茨，1991；阿尔梅达和费雷拉，2002）。这意味着，在一个企业平均规模较小或企业较多的地方，它的经济可能会更加稳定。一个在预计范围内的类似结果（一）集中决策过程不太善于管理冲突和成功。（二）分配的冲突损害效率调整（罗德瑞克，1999；阿尔梅达和费雷拉，2002）外生冲击。基于这些对比的考虑，一个重要的经验问题是，是否建立规模和产业结构等措施可能是系统地与经济表现的二阶矩措施相关。本文据此提出了初步的证据，关于选择工业结构和地方收入和就业增长率的波动性的相应措施之间的联系。我们使用的三种产业结构措施受到每个部门的广泛尊敬。正如谢弗（2

43、002），我们衡量平均的建立规模，通过员工人数、创造美元的附加值、发货或收据。我们也期待由两个对立因素推动的人均机构。寻找一份新的在某一有很多雇主的部门的工作应该更容易，它导致更多稳定水平的收入和就业增长率（即“就业选配假说”）。但是，其他条件不变，小企业往往会允许更多数量的企业并存，在这种情况下，在小企业就业（如上所述）记录的营业额往往会抵消许多公司更加稳定的利益。我们产业结构的每一个措施都被单独编译为制造业，批发，零售和服务行业。我们称这些结构措施为经济表现的两个二阶矩措施，即每年的人均实际收入增长率统计处和编制总额的年均增长速度的SD就业这些措施。这个结果导致了两个单独的文献股，经济增长

44、波动的实证协变量和建立规模与其他产业结构措施对宏观经济的影响。我们发现所有规模较大的企业和人均较多的部门都与稳定的就业增长速度相关联，符合营业额和就业选配假设。同样的联系是找到了，但并不带来所有部门人均实际收入增长速度的波动。下一节介绍了经验模型和样本。第三节介绍结果，而第四部分为结论。9二、模型及样本我们将关键变量嵌入一个标准的线性增长方程，XSY1其中Y是衡量经济表现，如上所述，是一个估计的截距项，S是产业结构衡量标准，因为上面的讨论，X是下面讨论的控制变量向量，和是估计系数，是随机误差项。如贝卡尔特等（2006），我们的经济增长率已经被测量了5年的时间。正如之前的经济增长的研究，控制矢量

45、包括人口的自然对数，平均土地面积的人口密度，教育和初始家庭收入中位数的衡量。人口是市场规模的测量措施在CETORELLI和GAMBERA（2001）。这也类似于总劳动人口的变量在OHUALLACHAIN和SATTERTHWAITE（1992）的使用，它可作为衡量城市化的经济解释。如果就业选配起来更快或者更有效地发生在人口较多的地区，那么对这个变量的估计系数应为负（另一个上面介绍的就业选配假说的含义）。人口密度已被发现与经济表现的几个第一时刻措施显著相关，可能是由于规模效应，或企业和工人之间的匹配，优越的密集市场（西科尼和霍尔，1996；安德森等，2004；卡诺等，2007；斯拉姆斯基等，200

46、5）。如果这些利益影响经济稳定，那么人们可能会想到估计人口密度系数应在我们的模型之中。教育是衡量人口年龄在25岁及以上的完成了高中学业的比例，并反映了人力资本积累水平。它类似于在拉詹和津加莱斯（1998），利文（2000年）和CETORELLI和GAMBERA（2001）等人先前的研究中经济增长的措施，并以平均增长研究利率理论之间的联系（特莱斯，2005）。一个关于教育的措施已被用于至少一年的生长波动研究（贝卡尔特等，2006）。此外，教育被认为是与美国大都会区的部门就业的OHUALLACHAIN和SATTERTHWAITE（1992）增长有关。如果教育促进经济稳定良好，其回归系数的估计应该是

47、否定的。家庭收入中位数可以初步反映经济表现的第一矩收敛作用的措施，如巴罗和萨拉伊马丁（1992）指出。同样的逻辑并不适用于二阶矩措施的经济表现，呈现这个变量的符号和系数的估计这一个开放性经验问题的意义。然而，类似的变量（初步的人均国内生产总值）已用于至少一年的生长波动之前的研究（贝卡尔特等，2006）。10表1总结了数据。我们的样本超过2000年的美国非大都市在测量19911995年的经济表现和1987年结构的措施。大都会地区排除在外，因为它们的边界一般不与个别县紧密联系，混淆从县级数据中得出的变量的测量问题。虽然没有在表中报告，两者的平均规模和人均设立机构之间的两两相关系数为021和018，

48、在批发行业和制造业中只有007和007。这些小型和高度的相关性表明，结构变量的两类反映了我们样本中独立统计的产业结构方面。性能数据的选择经确定的几年后，虽然随数据结构有助于降低反向因果关系的可能性，但共同所有的实证研究增长，因果关系不能最后确定。这种滞后结构也最大限度地减少潜在的内生性偏差，作为回归量是预先确定的。三、结果表2报告人均实际收入增长率标准差的回归估计，而表3报告就业增长率的SD估计。随着工业结构在三种方式上被四个部门的每一个测量，每个表报告12个回归。由于缺少或零数据，如表所显示在每个部门设立的几个县利用各种回归观测数字略有不同。显著性水平计算被从社会企业计算的异方差修正。在表2

49、中，各种结构的措施是统计在12个回归中的八个显著。研究发现，人均收入在制造业中更稳步增长，零售和批发单位雇用更多的工人，或者是由批发单位的年发货量测值较大。这些发现与上述讨论的营业额的假设是一致的。相反，稳定的增长被认为出现在雇用更少工人的批发单位或那些批发单位较少的城市或更多的人均制造场所。批发业的对比估计可能反映在非线性关系基础上超出了本研究范围的问题。这些结果表明，产业结构是系统相关的收入增长率的波动性的方式，在很大程度上是一致的，但在一些主要行业有所不同。表2中方程的拟合是温和的，调整范围2R从03到接近05。多变量控制大部分在001显著水平上。初步住户收入中位数是从来不显著的，相对于初步人均国内生产总值在贝卡尔特等（2006）研究中被发现的一些对消费增长波动性的回归。收入增长在人口较多的县更加稳定。人口密度和教育有助于提高人均收入增长率，虽然当控制人均批发场所时，教育会失去它的意义。教育的可变性与贝卡尔特（2006）等在越野范围中发现的相对比。在表3，各种结构措施在12个统计回归中的八个再次显著。在四个部门的任何一个有更多人均机构的地方，就业被看到的是更多的稳步增长，这与工作匹11配假说相一致。此外，稳定的就业增长发生在那些雇用很多工人的批发或服务机构，或在一个有较大的平均美元生意额的零售或服务的场所，这