1、原文EARNINGSQUALITYANDTHEPRICINGEFFECTSOFEARNINGSPATTERNSEARNINGSPATTERNSANDEARNINGSQUALITYPREVIOUSRESEARCHONPRICINGEFFECTSOFEARNINGSPATTERNSHASCONSIDEREDINCREASINGEARNINGS,EARNINGSTHATMEETOREXCEEDANALYSTFORECASTS,ANDSMOOTHEARNINGSTHERESEARCHISMOTIVATEDBYTHEOBSERVATIONTHATMANAGERSAPPEARTOFOCUSONMAINTA
2、ININGSUCHPATTERNSWEBUILDONTHISOBSERVATIONBYEXPLORINGTHENOTIONTHATMANAGERSACCRUALSCHOICESEITHERREINFORCEORUNDERCUTTHEPRICINGEFFECTSOFEARNINGSPATTERNSINTHISSECTION,WEDESCRIBEPREVIOUSRESEARCHSECTION21,DISCUSSWHYTHETHREEEARNINGSPATTERNSWECONSIDERMIGHTBEECONOMICALLYDISTINCT,ASOPPOSEDTOMANIFESTATIONSOFASI
3、NGLEUNDERLYINGCONSTRUCTSECTION22,ANDDISCUSSTHELINKSBETWEENEARNINGSPATTERNSANDEARNINGSQUALITYSECTION231、PREVIOUSRESEARCHONPRICINGEFFECTSOFEARNINGSPATTERNSBEGINNINGWITHTHEOBSERVATIONTHATMANYMANAGERSAPPEARTOSTRIVETOREPORTSTEADILYINCREASINGEARNINGS,BARTH,ELLIOTTANDFINN1999DOCUMENTREWARDSTOSUCHBEHAVIORFI
4、RMSWITHINCREASINGEARNINGSHAVEHIGHERPRICEEARNINGSMULTIPLESTHANOTHERFIRMS,AFTERCONTROLLINGFORGROWTHANDRISKTHECONTROLFORGROWTHISBASEDONBOTHINCREASESINBOOKVALUEOFEQUITYANDANALYSTEARNINGSGROWTHFORECASTS,ANDTHECONTROLFORRISKISTHEVARIANCEOFTHEMOSTRECENTSIXYEARSPERCENTAGEEARNINGSCHANGESTHEYATTEMPTTORULEOUTT
5、HEPOSSIBILITYTHATFIRMSWITHLONGPATTERNSOFINCREASINGEARNINGSALSOSHAREAVALUATIONRELEVANTFACTORTHATISKNOWNBEFORETHEEARNINGSPATTERNDEVELOPSRESULTSAREMIXEDALTHOUGHTHEAUTHORSCONCLUDETHATTHEWEIGHTOFTHEEVIDENCEDOESNOTSUPPORTTHEEXISTENCEOFSUCHAFACTORBARTHETALDONOTCONSIDERWHETHERTHEQUALITYOFEARNINGSMIGHTAFFECT
6、THEIRRESULTS,ALTHOUGHTHEYNOTETHATIFEARNINGSINCREASESAREOBTAINEDVIAEARNINGSMANAGEMENTWHICHINVESTORSDISCERNANDDISCOUNT,THEIRTESTSWILLBEBIASEDAGAINSTFINDINGPRICINGEFFECTSASSOCIATEDWITHPATTERNSOFINCREASINGEARNINGSSEETHEIRNOTE6KASZNIKANDMCNICHOLS2002ANDBARTOV,GIVOLYANDHAYN2002FOCUSONMARKETEFFECTSOFMEETIN
7、GOREXCEEDINGANALYSTFORECASTSKASZNIKANDMCNICHOLSDOCUMENTTHATCONSISTENTLYMEETINGOREXCEEDINGANALYSTFORECASTSISASSOCIATEDWITHAHIGHERMULTIPLERELATINGEARNINGSTOPRICE,ANDTHISPRICINGEFFECTISDUETOTWODISTINCTSUBEFFECTSHIGHERFUTUREEARNINGSANDAPREMIUMASSOCIATEDWITHTHEACTOFMEETINGOREXCEEDINGFORECASTSFORATLEASTTH
8、REEYEARSTHEYINTERPRETTHEIRRESULTSASINDICATINGTHATCONSISTENTLYMEETINGOREXCEEDINGANALYSTEXPECTATIONSCONNOTESLOWERRISKAND,THEREFORE,AHIGHEREARNINGSMULTIPLEHOWEVER,THEYALSONOTETWOOPENISSUESFIRST,THEYDONOTINVESTIGATEWHYINVESTORSWOULDASSOCIATELOWERRISKWITHCONSISTENTLYMEETINGANALYSTFORECASTSSECOND,THEYDONO
9、TINVESTIGATETHEROLEOFMANIPULATINGEITHEREXPECTATIONSOFEARNINGSOREARNINGSTHEMSELVESWEBELIEVEOURANALYSESSHEDLIGHTONTHESEISSUESSPECIFICALLY,WEPOSITTHATINVESTORSPRICEFIRMSTHATMEETOREXCEEDANALYSTEARNINGSEXPECTATIONSANDHAVEHIGHEARNINGSQUALITYHIGHERTHANFIRMSTHATMEETTHEFIRSTBUTNOTTHESECONDCONDITIONUSINGADIFF
10、ERENTRESEARCHDESIGNTHANKASZNIKANDMCNICHOLS,BARTOVETAL2002DOCUMENTHIGHERQUARTERLYABNORMALRETURNSINQUARTERSWHEREFIRMSMEETOREXCEEDFORECASTSSEVERALOFTHEIRTESTSDISTINGUISHBETWEENSHIFTSINANALYSTFORECASTSOVERTHEQUARTERANDTHEEARNINGSSURPRISE,BASEDONIBESREPORTEDEARNINGSCOMPAREDTOTHEMOSTRECENTIBESFORECASTTHEY
11、INTERPRETTHEFINDINGTHATTHERETURNTOANEARNINGSSURPRISEGREATLYEXCEEDSTHERETURNTOSHIFTSINANALYSTFORECASTSASRATIONALIZINGEXPECTATIONSMANAGEMENTTHEMARKETPENALTYFORDAMPENINGANALYSTEXPECTATIONSISSWAMPEDBYTHEREWARDTOAFAVORABLEEARNINGSSURPRISESUBSEQUENTTESTSSUPPORTTHISINTERPRETATIONMOSTDIRECTLYRELEVANTTOOURAN
12、ALYSES,BARTOVETALTESTWHETHERTHEPREMIUMFORMEETINGOREXCEEDINGANALYSTFORECASTSISSMALLERWHENTHEREISEVIDENCEOFEITHERORBOTHEXPECTATIONSMANAGEMENTANDEARNINGSMANAGEMENTTHEYDEFINETHEFORMERASTHECOMBINATIONOFNEGATIVEANALYSTFORECASTREVISIONSANDZEROORPOSITIVEEARNINGSSURPRISES,ANDTHELATTERASTHEPRESENCEOFUNEXPECTE
13、DACCRUALSBASEDEITHERONTHEJONES1991MODELASMODIFIEDBYDECHOW,SLOANANDSWEENEY1995ORONDEFININGNORMALACCRUALSASWORKINGCAPITALACCRUALSPLUSDEPRECIATIONANDAMORTIZATIONTHEYCONCLUDETHATBOTHEXPECTATIONSMANAGEMENTANDEARNINGSMANAGEMENTAREASSOCIATEDWITHSTATISTICALLYRELIABLEBUTSMALLDECREASESINTHEMARKETPREMIUMTOMEET
14、INGOREXCEEDINGANALYSTEXPECTATIONS,ANDTHATTHE“ECONOMICALLYMINOR”P198EFFECTOFEARNINGSMANAGEMENTMIGHTBEDUETOLOWPOWERWEEXTENDTHESEANALYSESBYPROVIDINGDIRECTTESTSOFWHETHERTHEPRICINGEFFECTSOFCONSISTENTLYMEETINGOREXCEEDINGANALYSTFORECASTSAREAFUNCTIONOFEARNINGSQUALITY,ANINVERSEMEASUREOFEARNINGSMANAGEMENTTHOM
15、ASANDZHANG2002REPORTHIGHERPRICEEARNINGSMULTIPLESFORFIRMSWITHSMOOTHEARNINGSTHEIRMEASURESOFEARNINGSSMOOTHNESSAREDERIVEDFROMTHESTANDARDERRORSOFREGRESSIONSOFEARNINGSONTIMEANDFROMTHESTANDARDDEVIATIONSOFSEASONALLYDIFFERENCEDQUARTERLYEARNINGSPERSHARETHEYCONDITIONONINTRINSICEARNINGSVOLATILITYPROXIEDBYTHEOVE
16、RTIMEVARIABILITYOFANALYSTEARNINGSFORECASTS,TOCAPTURETHEVALUATIONEFFECTSOFMANAGERIALINTERVENTIONSTOSMOOTHEARNINGSRESULTSINDICATETHATBOTHLOWERINTRINSICEARNINGSVOLATILITYANDEARNINGSSMOOTHNESSAREASSOCIATEDWITHHIGHERPRICEEARNINGSRATIOSANDTHATTHISRELATIONISDUEATLEASTINPARTTOANASSOCIATIONBETWEENLOWERRISKAN
17、DLOWERINTRINSICEARNINGSVOLATILITYLIKETHOMASANDZHANG,HUNT,MOYERANDSHEVLIN2000REPORTTHATEARNINGSMULTIPLESINCREASEWITHEARNINGSSMOOTHNESSTHEIRMEASURESOFSMOOTHNESSAREDERIVEDFROMASEPARATIONOFEARNINGSINTOCASHANDTOTALACCRUALS,ANDTOTALACCRUALSINTOITSDISCRETIONARYANDNONDISCRETIONARYCOMPONENTS,USINGTHEJONES199
18、1MODELASMODIFIEDBYDECHOW,SLOANANDSWEENEY1995CONTROLLINGFORTHESTANDARDDEVIATIONOFCASHFLOWSPRESUMABLYASAPROXYFORUNSMOOTHEDEARNINGSHUNTETALASSESSTHEINCREMENTALEFFECTSONPRICINGMULTIPLESWHEN1THESTANDARDDEVIATIONOFCASHPLUSNONDISCRETIONARYACCRUALSISSMALLRELATIVETOTHESTANDARDDEVIATIONOFCASHFLOWSAND2THESTAND
19、ARDDEVIATIONOFREPORTEDNETINCOMEISSMALLRELATIVETOTHESTANDARDDEVIATIONOFCASHPLUSNONDISCRETIONARYACCRUALSTHEFORMERMEASUREISINTERPRETEDASSMOOTHNESSINDUCEDBYNONDISCRETIONARYACCRUALSANDTHELATTER,ASSMOOTHNESSINDUCEDBYDISCRETIONARYACCRUALSHUNTETALCONCLUDETHATDISCRETIONARYSMOOTHNESSISASSOCIATEDWITHALARGERPOS
20、ITIVEPRICINGEFFECTTHANNONDISCRETIONARYSMOOTHNESS,ANDTHATDISCRETIONARYSMOOTHNESSISALSOASSOCIATEDWITHGREATEREARNINGSPERSISTENCETAKENTOGETHER,PRIORRESEARCHSUGGESTSTWOEMPIRICALREGULARITIESWHICHWEEXPLOREFIRST,THEREISACONSISTENTPOSITIVEPRICINGEFFECTASSOCIATEDWITHEARNINGSTHATARESMOOTH,EARNINGSTHATINCREASE,
21、ANDEARNINGSTHATMEETOREXCEEDANALYSTFORECASTSPREVIOUSRESEARCHHASCONSIDEREDTHESEPATTERNSONEATATIMEANDUNCONDITIONALLYWECONSIDERTHEEFFECTSOFEACHEARNINGSPATTERNCONDITIONALONTHEOTHERS,TOANALYZEWHETHERTHEEFFECTSAREINFACTDISTINCT,ASOPPOSEDTOVARYINGMANIFESTATIONSOFASINGLEUNDERLYINGFACTORSECOND,THEREISMIXEDEVI
22、DENCESUGGESTINGTHATTHEPRICINGEFFECTSOFEARNINGSPATTERNSMIGHTBECONDITIONALONWHETHEREARNINGSAREMANAGEDTHEEFFECTSOFEARNINGSMANAGEMENTAPPEARTOBESMALL,PERHAPSBECAUSEOFLOWPOWERTESTSWEPROVIDEDIRECTASSESSMENTSOFTHEEFFECTSOFEARNINGSQUALITYANINVERSEMEASUREOFEARNINGSMANAGEMENT2、RELATIONSAMONGEARNINGSPATTERNSAPA
23、TTERNOFINCREASINGEARNINGSSIMPLYIMPLIESTHAT,OVERTIME,EACHSUCCESSIVEEARNINGSNUMBEREXCEEDSTHATOFTHEPREVIOUSPERIODTHEREISNONECESSARYIMPLICATIONOF,ORCONFLICTWITH,SMOOTHNESS,MEASUREDEITHERINTERMSOFOVERTIMEEARNINGSVARIABILITYIE,EARNINGSMIGHTTRENDUPWARDEITHERSMOOTHLYORBYVARYINGAMOUNTSEACHPERIODORINTERMSOFLO
24、WREPORTEDINCOMEVARIABILITYRELATIVETOINTRINSICINCOMEVARIABILITYTHEREISALSONONECESSARYIMPLICATIONOFMEETING,ORNOTMEETING,ANALYSTEXPECTATIONSINTERMSOFMANAGEMENTSBEHAVIOR,EARNINGSINCREASESCANBEACHIEVEDBYSOMECOMBINATIONOFINTRINSICGROWTHANDEARNINGSMANAGEMENTVIAEITHERORBOTHACCRUALSMANIPULATIONSANDREALACTION
25、SDEPENDINGONMANAGEMENTSPLANNINGHORIZONANDTHEUNDERLYINGECONOMICSOFTHEFIRM,EARNINGSINCREASESCANBEPLANNEDANDMANIPULATEDOVERLONGPERIODSBYSYSTEMATICALLYBUILDINGANDTHENUSING“RESERVES”OFACCRUALSTHATIS,SOMEPORTIONOFUNMANAGEDCURRENTPERIODEARNINGSCANBEDELAYED,VIAACCRUALSMANAGEMENT,ANDINCORPORATEDINTOLATEREARN
26、INGS,ASNEEDEDHOWEVER,ONCEANEARNINGSNUMBERISANNOUNCED,ITBECOMESAFIXEDTARGETTHATMUSTBEEXCEEDEDIFTHEPATTERNOFINCREASINGEARNINGSISTOBEMAINTAINEDINCONTRAST,PATTERNSOFEXCEEDINGANALYSTFORECASTSCANBEACHIEVEDBYINTRINSICPERFORMANCE,EARNINGSMANAGEMENTANDEXPECTATIONSMANAGEMENTEVIDENCEOFEXPECTATIONSMANAGEMENTISP
27、ROVIDEDBY,FOREXAMPLE,BARTOVETAL2002ANDMATSUMOTO20022TOTHEEXTENTMANAGERSCANANDDOGUIDEANALYSTSTOWARDAFORECASTTHATWILLENSUREAZEROORPOSITIVEEARNINGSSURPRISE,THEOVERTIMEPATTERNOFEARNINGSTHATMEETOREXCEEDANALYSTFORECASTSCANTAKEANYFORMOFINCREASESANDDECREASES,CANEXHIBITANYLEVELOFSMOOTHNESS,ANDCANBEENTIRELYUN
28、MANAGEDTHUS,THEREISNONECESSARYRELATIONBETWEENMEETINGOREXCEEDINGANALYSTFORECASTSANDANYOTHEREARNINGSPATTERN,ORBETWEENTHISEARNINGSPATTERNANDEARNINGSQUALITYFINALLY,EARNINGSSMOOTHNESSISTYPICALLYASSESSEDBYCALIBRATINGEARNINGSVARIABILITYOVERTIME,RELATIVETOTHEVARIABILITYOFSOMECONSTRUCTTHATISPRESUMEDTOREPRESE
29、NTUNMANAGEDANDUNOBSERVABLEFUNDAMENTALSFOREXAMPLE,THOMASANDZHANG2002USETHEOVERTIMEVARIABILITYOFANALYSTFORECASTSASAMEASUREOFINTRINSICVOLATILITY,WHILEHUNT,MOYERANDSHEVLIN2000EXAMINETHERATIOOFTHESTANDARDDEVIATIONOFOPERATINGCASHFLOWSPLUSNONDISCRETIONARYACCRUALSTOTHESTANDARDDEVIATIONOFOPERATINGCASHFLOWSOU
30、RMEASUREOFEARNINGSSMOOTHNESS,THESTANDARDDEVIATIONOFINCOMEDIVIDEDBYTHESTANDARDDEVIATIONOFOPERATINGCASHFLOWS,ISRELATEDTOHUNTETALSMEASURETHISRATIOCONTROLSFORUNDERLINGCASHFLOWVARIABILITY,ASLONGASTHERELATIONISPROPORTIONALIE,ASLONGASTHEMARGINALRELATIONISTHESAMEASTHEAVERAGERELATIONBECAUSEEARNINGSSMOOTHNESS
31、ISMEASUREDRELATIVETOAPROXYFORINTRINSICEARNINGSVOLATILITY,WHICHINTURNISDETERMINEDBYBUSINESSFUNDAMENTALSANDTHEECONOMICENVIRONMENT,THEREISNONECESSARYRELATIONBETWEENSMOOTHNESSANDANYOTHEREARNINGSPATTERNHOWEVER,EARNINGSSMOOTHNESSMIGHTBERELATEDTOPATTERNSOFEARNINGSINCREASESSIMPLYBECAUSEBOTHCANBETHERESULTOFA
32、CCRUALSMANAGEMENT3、RELATIONSBETWEENEARNINGSQUALITYANDEARNINGSPATTERNSASNOTEDEARLIER,WEMEASUREEARNINGSQUALITYINTWOWAYSTHEFIRSTMEASUREISDERIVEDFROMDECHOWANDDICHEVS2002REGRESSIONMODELRELATINGCURRENTACCRUALSTOCURRENT,PASTANDFUTURECASHFLOWSEARNINGSQUALITYISDECREASINGINTHESTANDARDDEVIATIONOFTHEFIRMSPECIFI
33、CREGRESSIONRESIDUALS,THATIS,INTHEEXTENTTOWHICHVARIATIONINACCRUALSISNOTEXPLAINEDBYVARIATIONINCURRENT,PASTANDFUTURECASHFLOWSTHISMEASUREOFEARNINGSQUALITYDOESNOTDISTINGUISHBETWEENMANAGEMENTINDUCEDNOISEINTHEACCRUALSCASHRELATIONANDNOISETHATISINTRINSICTOTHEFIRMSBUSINESSMODELANDECONOMICENVIRONMENTTHESECONDE
34、ARNINGSQUALITYMEASUREISTHEABSOLUTEVALUEOFPERFORMANCEMATCHEDABNORMALACCRUALS,ESTIMATEDFROMTHEJONES1991MODELASMODIFIEDBYDECHOW,SLOANANDSWEENEY1995PERFORMANCEMATCHINGFOLLOWSKOTHARI,LEONEANDWASLEY2002THISQUALITYMEASUREISINTENDEDTOIDENTIFYEARNINGSMANAGEMENT,BYSEPARATINGNORMALACCRUALSTHATARERELATEDTOACCOU
35、NTINGFUNDAMENTALSFROMABNORMALACCRUALSINTRODUCEDBYMANAGEMENTTOACHIEVESPECIFICEARNINGSOUTCOMESTHETWOEARNINGSQUALITYMETRICSAREINTENDEDTOCAPTURECONSTRUCTSTHATAREONLYPARTIALLYOVERLAPPINGIE,THEDECHOWDICHEVBASEDMODELCAPTURESBOTHOPPORTUNISTICACCRUALSANDHONESTESTIMATIONERRORSANDTHEJONESBASEDMODELWASDEVELOPED
36、TOCAPTUREONLYTHEFORMERWEPOSITTHATINVESTORSUSEEARNINGSQUALITYASACONDITIONINGVARIABLETOASSESSTHEEXTENTTOWHICHAGIVENOBSERVEDEARNINGSPATTERNISINTRINSICANDSUSTAINABLEWEARGUETHATIFEARNINGSARELINKEDTOACCOUNTINGFUNDAMENTALSWHICHWEMEASUREBYCURRENT,LASTPERIODANDNEXTPERIODCASHFLOWSINTHEDECHOWDICHEVQUALITYMETRI
37、CANDBYREVENUESANDFIXEDASSETSINTHEMODIFIEDJONESQUALITYMETRICTHOSEEARNINGSAREWELLSUPPORTED,HENCESUSTAINABLE3FORANYOBSERVEDPATTERNOFREPORTEDEARNINGS,ADISCONNECTBETWEENEARNINGSANDFUNDAMENTALSIMPLIESTHATTHEPATTERNISATLEASTINPARTILLUSORYINTHECASEOFEARNINGSINCREASESWEPOSITTHATLOWEARNINGSQUALITYIMPLIESTHATG
38、ROWTHINEARNINGSDERIVESMOREFROMNOISEINTHEACCRUALSACCOUNTINGFUNDAMENTALSRELATIONTHANFROMECONOMICFUNDAMENTALSINTHECASEOFMEETINGOREXCEEDINGANALYSTFORECASTS,WEPOSITTHATLOWQUALITYEARNINGSIMPLIESTHEUSEOFACCRUALSMANAGEMENTTOMEETTHETARGETHOWEVER,WEALSONOTETHATAPATTERNOFMEETINGOREXCEEDINGANALYSTFORECASTSISDIS
39、TINGUISHEDFROMOTHEREARNINGSPATTERNSBECAUSEITMIGHTBEACHIEVEDWITHOUTANYEARNINGSQUALITYEFFECTSWHATSOEVERFINALLY,INTHECASEOFSMOOTHEARNINGS,WEPOSITTHATLOWQUALITYEARNINGSIMPLIESTHATMANAGEMENTISUSINGACCRUALSTODRIVEAWEDGEBETWEENUNMANAGEDACCOUNTINGFUNDAMENTALSANDREPORTEDEARNINGSFOREACHEARNINGSPATTERN,WEPREDI
40、CTTHATLOWQUALITYEARNINGSWILLDIMINISHOREVENELIMINATETHEPRICINGEFFECTSOTHERWISEASSOCIATEDWITHTHEEARNINGSPATTERNSOURCEJENNIFERFRANCIS,RYANLAFOND,PEROLSSON,KATHERINESCHIPPER,ACANEWS,APRIL2003,PP2530译文盈余质量与盈余的定价模式的影响盈余模式和盈余质量以往的研究对于盈利模式的定价影响已经考虑增加盈利，使盈利达到或超过分析师预测，并顺利的收益。这项研究是以观察为动力的，管理者似乎把重点放在维持这样的模式。在这种
41、观察的基础上，我们探索管理者选择加强应计收入或者削弱定价模式的意图。在本节中，我们描述过去的研究并且讨论为什么我们认为三盈利模式可能在经济上不同，而不是一个单一的表现形式的基础建设，并讨论盈利模式之间的联系和收益质量。1、过去的研究针对的盈利模式定价的影响在观察刚开始的时候，许多管理人员似乎努力的报告不断增加的收入，巴特，埃利奥特和芬兰1999文件奖励这种行为在市场增长和风险的控制后，收入增加的企业比其他企业有更高的市盈率倍数。市场增长的控制是基于账面价值的股票和分析师账面盈利增长预测值都增加这两点上的，而风险控制是基于最近六年的收益百分比差异变动情况。他们试图排除长期模式盈利增加的企业也都有
42、一个相关的因素是估值的盈利模式发展之前已知的这种可能性结果是复杂的，尽管作者的结论没有强有力的证据支持这种因素的存在。巴特等人不考虑收益质量是否会影响他们的结果，尽管他们注意到如果收益增加是通过投资者辨别真伪和获得折扣的盈余管理机构公布的，那么他们的测试将会受到盈利模式的定价影响而产生偏差。KASZNIK和麦克尼科尔斯2002和BARTOV，GIVOLY和HAYN2002关注于达到或超过分析师预测的市场效应。KASZNIK和麦克尼科尔斯证明始终达到或超过分析师预测是与高度的共享盈利价格相关得，而且这种价格效应是由于两种不同的子效应更高的未来收益和一种在3年内达到或者超过预测的额外收益。他们的结
43、果表明，始终达到或超过分析家的预期蕴含的风险较低，因此，有较高的市盈率。不过，他们也注意到两个悬而未决的问题。首先，他们不调查为什么投资者一直维持较低的风险而能始终达到分析师的预测。第二，他们不调查操纵自己的收入或盈利或预期的作用。我们相信，我们的分析很清楚的揭示了这些问题；具体来说，我们假设企业投资者的价格，达到或超过分析师的盈利预期，而且比其他公司具有较高的盈利质量是首要条件而不是次要条件。使用一个不同的研究设计比KASZNIK和麦克尼科尔斯，BARTOV2002文件在那些达到或者超过预期的公司有更高的异常回报。根据IBES报道的盈利和他们预测的盈利的比较，他们的一些测试区分季度收入预测和
44、意外收入的变化。他们解释发现意外收入的回报远远超过了分析师的预测变化收入。随后的测试也支持了这种解释。最直接和我们的分析相关的是BARTOV等测试当有一方或者双方都有预期的管理与盈余管理的证据时，是否达到或超过分析师预测的溢价较小。它们定义前者为消极的分析师修改预测和零和真实的意外收入的结合，作为后者存在意想不到的应计费用取决于琼斯模型的修改周期，斯隆和斯威尼或营运资金应计项目定义为加折旧及摊销正常应计项目。他们的结论是，预期的管理机构和收益管理机构都与可靠的统计有关，但市场的跌幅较小，议价达到或者超过分析师的预期，而且认为“轻微经济”（第198页）的盈余管理的影响可能是取决于低损耗。我们通过
45、测试价格影响中向来达到或者超过分析师的预测是否是盈利质量的作用来深入分析，作为盈余管理的一个反面措施。托马斯和张2002报告高的市盈率倍数的公司有稳定的盈利。他们稳定盈利的措施是及时收益回归的标准误差和每一股不同季度的季度性盈利的标准差。他们取决于盈利波动的内在因数（由分析师盈余预测逾时的变化为代表），来完成他的管理干预效果评价的评估。结果表明，内在的盈利波动性和收益平滑性都是和较高的市盈利定量有关，而且这种关系部分取决于低的风险和低的内在的盈利波动性。托马斯，张，亨特分析在小市场溢价下降达到或超过分析师预期时，他们认为“轻微经济”（第198页）的盈余管理的影响可能是由于低损耗引起的。我们通过
46、提供其是否始终达到或超过分析师预测的价格影响来分析收益质量，并且预期其收益成果像托马斯，张，亨特，莫耶和谢夫林2000报告的一样，市盈率的增长与盈余平稳有关。他们平稳的措施是分离现金收入和总应计，而且总应计酌情或者非全权纳入，利用由DECHOW，斯隆和斯威尼1995的琼斯模型1991修改。控制现金流量为标准差，这应该是评估价格的倍数时的增效作用（1）现金加NONDISCRETIONARY预提费用的标准差相对较小，现金流量标准差（2）标准差的净所得相对较小的现金加上标准偏差NONDISCRETIONARY应计费用。前者的措施是被解释为由非可操作应计利润和后者产生的平滑度，作为可操作应计利润诱导的
47、平滑度。亨特等人。得出结论，酌情平滑与一个大于NONDISCRETIONARY平滑积极定价的效应，而且也酌情平滑更大收益PERSISTENCELIKE托马斯和张，亨特，在市场溢价达到或超过分析师预期小减小，而相关在“经济小”（第198页）的盈余管理的影响可能是由于低功耗。我们通过提供扩展是否始终达到或超过分析师预测的价格影响是收益质量，盈余管理的一个反措施，这些分析功能的直接考验。托马斯，张，亨特，莫耶和谢夫林的报显示市盈率与盈余平稳增长。其措施是来自一成现金，总应计项目，总应计收入纳入其自由裁量和NONDISCRETIONARY成分分离，利用由DECHOW，SLOAN和斯威尼修改的琼斯模可见
48、控制现金流量为标准差（这应该是平滑的收益为代理）。评估价格的倍数时增效作用（1）现金加NONDISCRETIONARY预提费用的标准差相对较小，现金流量标准差（2）标准差的净所得相对较小的现金加上标准偏差NONDISCRETIONARY应计费用。前者的措施是由NONDISCRETIONARY解释为，而后者引起的应计光滑，经裁决性应计诱导畅通。亨特等人。所以我们得出结论，酌情平滑与一个大于NONDISCRETIONARY平滑积极定价的效应，这两者与酌情平滑盈余持续性有很大的关联。两者合计，以前的研究表明，探索我们两个经验的规律性。首先，有一个一致的积极的平滑盈利定价效应，从而能使获利增加，企业盈
49、利达到或超过分析师的预期。此前有研究认为，这些在时间和无条件模式之中存在。我们认为，每一个盈利模式的影响，对别人的条件，分析是否影响其实是不同的，而不是不同的表现形式单一的根本因素。第二，有证据表明，盈利模式的影响可能是定价是否盈利的管理条件。盈余管理的影响似乎要小，也许是因为低损耗试验。对我们提供的收益质量（一种盈余管理的反措施）影响的直接评估。2、盈利模式间的关系一种简单增加收入的模式意味着，随着时间的推移，每个数字连续获利超过了去年同期。没有任何必然含义或冲突，平滑地衡量收益随着时间的推移而变化（即无论收益可能趋势平稳或不同程度的每一时期都会向上递增）对于低报收入得变异性，无论是相对于内在收益变性还是外在收益的变异性，都不符合分析师得预期。在管理的行为方面，盈利的增加，可达到某种内在的增长和盈利管理相结合的状态。一方或双方应计通过操纵和实际行动来开拓管理层的规划视野。该公司的基本经济学的不同，导致收入的增加可以计划和有系统的建设，然后使用“保留”的预提费用在长期内操纵经济的发展，也就是一些非托管当期收益部分可延迟，通过预提费用的管理，并纳入到后来的收入。根据需要，一旦盈利人数出炉，就成了一个固定的目标，必须超过收入的增加，当然是在要维持模式的前提下。与此相反，超过分析师预测的模式可以通过内在性能，盈余管理和预期的管理而变化。预期管理提供的证据，例如，BARTOV
