股利政策：爭議問題外文翻譯

原文：Dividendpolicy:TheIssues1.IntroductionThedeterminantsofdividendpolicyareacontinuingpuzzle,asnotedbyBlack(1976).Inthispaperwereviewthemajorissuesindividendpolicyandrelatethemtosomeofthethemesexploredincompanionpapersinthisvolume.Thepaperisdividedintofivesections.Section2surveystheliteratureontheinformationsignallingpropertiesofdividends.Section3discussessometaxissuesrelatedtodividendpolicyandsection4drawsonsomeagencycostsexplanationsfordividendpayments.Theconclusiondrawstogethertheargumentsandhighlightssomeoftheunresolvedissues.2.DividendpolicyandinformationsignallingIntheirclassicpaper,MillerandModigliani(1961)provideacogentargumentfortheclaimthatdividendpolicydoesnotaffectthevalueofthefirm.Theyassumeaworldwithouttransactionscostsandtaxes,agiveninvestmentpolicyandfullyinformedinvestors.Inthesecircumstancesitfollowsthat"theirrelevanceofdividendpolicy,giveninvestmentpolicy,is'obvious,onceyouthinkofit'"(p.414).Withagivenlevelofinvestment,ifafirmchoosestopayadollarmoreofdividendnow,itwillhavetoraiseanextradollarofexternalfinancetosupportitsinvestment:Thehighercurrentdividendtoexistingshareholderswillbeexactlyoffsetbyadecreaseinfuturedividendsasthefirmmustnowpaydividendstoitsnewshareholders.Intheabsenceoftaxeffectsandtransactionscostsandgivenfullinformation,thevalueofthefirmtoexistingshareholderswillnotbeaffectedbyitsdividendpolicy.Dividendpolicyisthereforeirrelevant.MillerandModigliani(1961)notethatthe"informationalcontent"ofdividendsisassumedabsentfromtheirmodel.Theynote,however,thatinpractice"whereafirmhasadoptedapolicyofdividendstabilizationwithalong-establishedandgenerallyappreciated'targetpayoutratio',investorsarelikelyto(andhaveagoodreasonto)interpretachangeindividendrateasachangeinmanagement'sviewsoffutureprofitprospectsforthefirm".ThisechoedthestancerecordedbyLintner(1956)inhissurveyofU.S.companypracticeswhichledhimtodevelopthe'partialadjustment'modelofdividendbeha-viour.Lintnersuggeststhattheprimarydeterminantofdividendpaymentsistherelationshipbetweentheexistingdividendrateand"thatratewhichwouldconstituteatargetpay-outofcurrentandreasonablyforseeableprofits".AnumberofdifferentinterpretationsofLintner'sbehaviouralmodelarepossible.Oneinterpretationisthatmanagementsignalschangesintheirperceptionofthecompany'slong-termprofitabilitybychangesindividendpayments.ApaperbyAllen,inthisvolume,assessestheuseoftargetpayoutratiosbyasampleofBritishcompanies,drawingonevidenceobtainedinaquestionnairesurvey.TheresultsareconsistentwiththeimplicationsoftheLintnermodel.Respondentsreportthatadesiretomaintainstabledividendsandthecompany'srecentdividendhistoryarethemainfactorsinfluencingtargetpayouts.Theimportanceoftheuseofpayoutlevelstosignalchangesinexpectationsoffutureearningsisalsoemphasized.Theroleofinformationsignallinghasbeenemphasizedinrecenttheoreticalworkondividendpolicy.MillerandRock(1985)provideaformalmodeloftheroleofdividendpolicyunderasymmetricinformation.Theyassumethatfirm'smanagersknowmorethanoutsideinvestorsaboutthetruestateofthefirm'scurrentearnings;bothdividendpaymentsandexternalfinancingwillconsequentlyhaveannouncementeffectsastheyprovideinformationaboutthefirm'ssourcesandusesoffundsthatwillenablethemarkettodeducethefirm'scurrentearnings.Thisestimateofcurrentearningscanthenbeusedtoestimateexpectedfutureearningsandfirmvalue.Unfortunately,thisleadstothepossibilitythatmanagementmaytrytopushupthesharepriceintheshort-runbyover-generousdividendpayments.Ultimately,thetruthwillberevealedbutnotbeforesomeinvestorshavereapedanexcessreturnbysellingbeforethetrueinformationisrevealed.MillerandRocksuggestapotentialsolutiontothisproblembutattheexpenseofFisher'scriterionforoptimalinvest-ment.Theyprovideasignallingequilibriuminwhichthedividendpayoutratioishigherandrealinvestmentislowerthaninthefullinformationcase.Theaboveapproachisconsistentwiththewell-documenteddividendan-nouncementeffectsasreportedbyAharonyandSwary(1980),AsquithandMullins(1983),andBrickley(1983).BallandBrown(1968)providethefirstevidenceofthelinkagebetweenunexpectedchangesinearningsandshareprices.3.TaxationanddividendpolicyThereareanumberofpotentiallinkagesbetweencorporatedividendpoliciesandthetaxsystem.FarrarandSelwyn(1967)andBrennan(1970)pointout,thatintaxregimeswheredividendincomeistaxedatahighermarginalpersonaltaxratethancapitalgains,shareholdersindifferenttaxbracketswillpreferdifferentpayoutpolicies.Itmakessenseforhigherpersonaltaxrateinvestorstoinvestinlowerdividendpayoutstockssothattheycantakemostoftheirgainsintheformofcapitalgains.Indeed,pushedtothelimit,thisargumentsuggeststhatitistaxinefficient,fromahighlytaxedinvestor'spointofview,foracompanyinthistypeoftaxregimetopayanydividends.Ifequilibriumconditionsareassumed,theargumentalsosuggeststhatrationalinvestorsshoulddemandahigherrisk-adjustedreturn,onapre-taxbasis,fromhighdividendyieldstocks.Thiswillcompensatefortheirrelativelyunfavourablepersonaltaxtreatment.Trueman(1986)modeltheinteractionofdividendpolicy,invest-mentdecisionsandfinancingdecisionsinthecontextofacorporatetaxregimewithvaryinginvestorpersonaltaxrates.Theirmodelimpliesthatshareholdersindifferentpersonaltaxbracketswillnotagreeonwhatconstitutestheoptimalinvestment/divi-dendpolicy.Investorsinhighpersonaltaxbracketswouldpreferthefirmtoinvestmorewhilstthoseinlowertaxbracketswillpreferhigherpayoutsandlowerinvest-ment.Thisproblemwillbereducedifclientelesofinvestorsclusteraroundfirmswithdifferentdividendpayoutpolicies.ThepossibilityofadividendclienteleeffectwasfirstsuggestedbyMillerandModigliani(1961).Itprovidesonepotentialexplanationofcompanies'observedreluctancetoaltertheirdividendpayoutratios.Theresultwouldbethatshareholderswouldincurtransactionscostsinrearrangingtheirportfoliostoachievedesiredincomestreams.EltonandGruber(1970)examinethedrop-offratio,definedasthedifferencebetweentheex-dividendandcum-dividendpricescaledbythedividend,totryandmeasureclienteleeffects.Ifthepricedeclinesbythefullamountofthedividend,afterthepaymentofthedividend,thedrop-offratioisone.EltonandGruberreportapositiveassociationbetweenthemeandividendyieldsanddrop-offratios.Theyinterpretthisasbeingconsistentwiththeexistenceofataxclienteleeffectandinferthemarginaltaxratesofinvestorsfromthesizeofthedrop-offratio.Theimplicationisthatshareswithalargerdrop-offratiopossessaclienteleofinvestorswithalowertaxburdenondividendsreceived.Kalay(1982)hasarguedthatitisnotpossibletoinferinvestorclienteletaxratesfromthesizeofthedrop-offratio.Thisisbecausedrop-offratioswouldbeboundedbytransactionscosts,notmarginaltaxrates.ApaperbyClarke,inthisvolume,assessestheimpactofachangeintaxregimeondrop-offratiosinasampleofAustraliansharesresultingfromtheswitchtoadividendimputationtaxsysteminJuly1987.However,thesedifferenceswerenotstatisticallysignificantandthelargeamountofnoiseinthedatamakesitdifficulttodrawclearcutconclusions.FurtherevidencerelatingtotheimpactofchangesindividendtaxationregimesonsecuritypricesandyieldsisprovidedbyPoterbaandSummers(1984).TheyanalyseasampleofBritishsharepricedataforaperiodfrom1955to1981.4.AgencycostsanddividendpolicyJensen(1976)demonstratethatifmanagementservetheirowninterestsandnotthoseofoutsideshareholders,thenagencycostscouldarise.Thesewillresultbecauseshareholdersforseethatmanagerscanincreasetheirownwealthattheirexpensebytheexcessiveuseofperksorbyshirking.IntheJensenandMeckling(1976)scenario,owner-managersmayfinditoptimaltoincurmonitoringandbondingcoststoreducepotentialagencycosts.Theseagencycoststakeanumberofformsincludingmonitoringcostsandexcessiveriskaversionbymanagerswhohaveasignificantportionoftheirownwealthtiedupwiththefortunesofthefirm.Undulyconservativebehaviourmayservetheinterestsofthefirm'screditorsbutnottheshareholders.Rozeff(1982)suggeststhatintheabsenceoftaxes,itispossibleforafirmtohaveanoptimaldividendpolicyduetotheexistenceofagencycosts.Hearguesthatdividendpayoutratioscouldbeconditionedbyatradeoffbetweentheflotationcostsofraisingexternalfinanceandthebenefitofreducedagencycosts,realisedwhenthefirmincreasesitsdividendpayoutratio.Easterbrook(1984)alsosuggeststhatdividendpaymentsmayservetoreduceagencycosts.Managementcanchangetheriskofthefirmbychangingboththeprofileofitsrealinvestmentprojectsandtherelativebalanceofdebtandequity.Bymaintainingaconstantpaymentofdividendsitavoidsabuildupinthebalanceofequityfundsandsimultaneouslyforcesthefirmtoseekexternalfinance.Theraisingofexternalfinancewillcauseperiodicreviewsofthefirm'sactivitiesbythecontribu-torsofcapital;theirpresencethereforeeasestheburdenonexistingshareholders.Shareholderswillalsobenefitfromtheadjustmentofleverageratioswhichwillaccompanytheraisingofexternalfinance.Thisargumentalsoprovidesapotentialexplanationofwhycompaniespaydividendsandraiseexternalfinanceatthesametime.Thefinalpaperinthisvolume,byAllen,examinessomeofthepecularitiesofJapanesecompanydividendpolicies.Japanesedividendpolicieshaveanumberofinterestingaspects.Thestandardpolicyistopayadividendwhichis10percentoftheparvalueofthestock.Atfirstglance,thissuggeststhatdividendpolicycannotactasasignallingdeviceinJapan.However,thepaymentofspecialandmemorialdividendsiscommonandthereisanemphasisonstocksplits,bothofwhichcouldbeusedassignallingmechanisms.Ontheotherhand,manyJapanesecompaniesaremembersofgroupswithstablesystemsofcross-shareholdings.Thegroupbanksholdequityingroupcompaniesaswellasprovidingloans.Asgroupcompanieshavemanypointsofcontactandchannelsofinformation,theyareunlikelytorequireinformationsignalsfromdivi-dendpayments.Furthermore,thefactthatmajorlendersarealsoshareholdersmeansthatconflictsofinterestgivingrisetoagencycostsarealsolikelytobereduced.DividendpolicyshouldhavelessofaroleinreducingagencycostsinJapan.Indeed,itisquiteconceivablethattherearethreedistinctcategoriesofshare-holdersinJapan:stablegroupshareholders,externalinstitutionalinvestors,andexternalindividualshareholders.Institutionalinvestorswouldconceivablybenefitfrommoregenerousdividendpayoutpoliciestohelpthemmeettheirliabilities.AccountingstandardsandpracticesplusmethodsofconductingcompanyannualgeneralmeetingsmeanthatJapaneseindividualinvestorsarenotlikelytobewell-informedaboutcompanyprospectsrelativetostableshareholderswhoshouldbemuchbetterinformed.Dotheformergleananyinformationfromdividendpolicy?ItisnotclearwhyJapanesecompaniesfollowtheircurrentdividendpolicies,norisitclearwhethertheycanservetheinterestsofalltheirstakeholdersequally.5.ConclusionDividendpolicyremains,asBlack(1976)suggested,"apuzzle".TheevidenceintheWestseemstosuggestthatdividendpaymentsdoimpartinformation.Thereisalsosomeevidenceoftaxationimpactinguponsharepricesandreturnsviadividendpolicies.Whydofirmspaydividendsandraiseexternalfinanceatthesametime?ItmaybethecasethatWesterncompaniesseesomebenefitintheformofreducedagencycostsresultingfromthejointimpactofthepolicy.However,ithastobeadmitted,dividendpoliciespursuedinJapandonotfitneatlywithanyoftheabovetheories.Source:D.E.Allen,H.Y.Izan,(1993)"DividendPolicy:TheIssues",ManagerialFinance,Vol.18Iss:1,pp.1–8.二、翻譯文章譯文：股利政策：爭議問題1、介紹股利政策的決定因素是一個持續(xù)的謎題,值得注意的是由布萊克(1976)在文中提出的論點,在文中我們回顧股利政策的重大課題以及關(guān)注了一些與股利政策相關(guān)的文章。論文分成了五個部分。第二部分調(diào)查的文獻是信息信號性能的股息。第三部分討論一些股利政策的相關(guān)的稅收問題，第四部分引用了一些代理成本股利發(fā)放的解釋。結(jié)論得出的這些觀點一起強調(diào)了一些懸而未決的問題。2、股利政策和信息發(fā)送信號米勒和莫迪里阿尼（1961）在他們的經(jīng)典論文中提供了一個有力的論據(jù)聲稱股利政策不影響企業(yè)的價值。他們假設(shè)世界上沒有一個全面的交易成本和稅收，給定的投資政策和投資者有的充分信息。在這樣的情況下,遵循“毫不相干的股利政策”,“一旦你認為是”對于投資政策的作用是明顯的。在給定的投資水平下,如果一個公司現(xiàn)在選擇支付一美圓更多的股息,它必須籌集外部資本額外的一美元以支持他們的投資：更高的的紅利正好抵消現(xiàn)有股東未來股利的減少和企業(yè)現(xiàn)在必須支付股息給新股東。在缺乏稅收效應(yīng)和交易成本、信息充分下企業(yè)的價值對現(xiàn)有股東不受股利政策影響。因此，股利政策是無關(guān)。米勒和莫迪利亞尼（1961）注意到，“信息內(nèi)容”股息假設(shè)偏離了他們的模型。這呼應(yīng)林特納（1956）在他的調(diào)查中美國公司的活動中使他產(chǎn)生了用不同的模型部分調(diào)整的股息的記錄。林特納表明，發(fā)放股利的主要決定因素是現(xiàn)有的股息率的存在和“這個速度即會構(gòu)成目標的當(dāng)前和合理可預(yù)見的利潤”對于這點林特納有不同的解釋可能。一種解釋是,公司的長期贏利和股利發(fā)放通過改變管理信號的感知變化而變化。艾倫在文章中評估了對通過借鑒英國公司進行問卷調(diào)查利用了目標的支付比率的樣品證據(jù)。其結(jié)果比較吻合林特納模型的含義。受訪者報告說，一個希望保持穩(wěn)定分紅和公司最近股息歷史是目標支付的主要因素影響。使用的重要性支出水平變化信號對股票未來盈利的預(yù)期也是強調(diào)的重點。信息的信號的作用也是最近股息政策理論一直強調(diào)的。米勒和羅克（1985）提供一個正式模式作用的非對稱信息條件下的股利政策。他們認為公司的管理者了解公司目前的盈利超過了對外部投資者了解的真實情況；同時紅利收支和對外融資將因此需要發(fā)布相關(guān)的影響消息，因為它們提供資金的信息是有關(guān)公司的消息來源和用該將使市場來推斷公司的當(dāng)期損益。目前的估計，這收益可以被用來估計預(yù)期未來收益和公司價值。不幸的是，這將導(dǎo)致管理可能會嘗試試圖太高股價短期過度支付高股利。最終，真相將被揭示而不是之前一些投資者賺取超額返回銷售前的真實信息顯示出來。米勒和羅克提出一個潛在解決這個問題，但在代價是Fisher的最佳投資標準。它們提供一個較高的信令均衡的派息率,實際利用外商直接投資是低于完整信息的情況。保爾和布朗（1968）提供了第一個證據(jù)是著名的對利潤和股票價格之間意想不到變化的聯(lián)系觀點。3稅務(wù)及股息政策在股利政策和稅收制度有一些潛在的聯(lián)系。法拉和塞爾溫（1967）和布倫南（1970）指出，法勒和塞爾溫(1967)和布倫南(1970)指出,在稅收制度中在股息收入是在更高的邊際個人所得稅資本收益中得到的,股東的比例比在不同的稅級下將喜歡不同的支付政策，這是有道理的，稅率更高的個人投資者投資于低派息股票以便他們能夠獲得他們大部分的獲取形式的資本收益。事實上，是推到了極限，這種說法表明，它是稅收效率不高，從高度征稅投資者的角度來看對于公司一個因為在這類型的稅收制度愿意支付任何股息。如果條件公平,爭論結(jié)果也表明,合理的投資者應(yīng)該要求更高的風(fēng)險調(diào)整后的回報,在稅前股票高股息率。這將彌補各自的相對不利的個人稅收待遇。特魯曼（1986）模型的政策之間的相互作用分紅，投資決策和融資決策的制度下的法人稅與不同投資者的個人所得稅稅率。他們的模型意味著在不同的股東個人所得稅不會同意什么構(gòu)成了最優(yōu)投資/股息的政策表示的公司股息政策。投資者在高個人稅級下寧可在公司同時進行更多的投資在低稅率下會選擇更高的支付和較低的投資。米勒和莫迪里阿尼(1961)首次提出這個問題將造成客戶集群減少，投資者在公司不同的派息政策下可能影響顧客的紅利。它提供了一個人可能的解釋是觀察到公司不愿改變其股利支付率。其結(jié)果將是股東會招致達到預(yù)期的交易成本而重新安排他們的投資組合，以達到預(yù)期的收入來源。埃爾頓和格魯伯（1970）研究了下降率，差定義除權(quán)和附帶股息分紅價格比例，試圖衡量對客戶的影響。如果價格的下滑,由于全額的紅利,紅利付款的比例下滑。埃爾頓和格魯伯報告一個積極關(guān)聯(lián)的意思是股息收益率和下滑率