speeches · February 10, 2008

Speech

William Poole · President

Reflections TheSt.LouisGatewayChapteroftheNationalAssociationforBusinessEconomics St.Louis,Missouri February11,2008 Ihave now attended my last FOMC meeting individualpolicyactionstotake?Tolearnfrom and, once editing of the meeting minutes experience,wemustbeabletogeneralizeand is complete, I will no longer be a part of determinewhatcircumstancescallforwhat the FOMC process. I am in a transition actions. period, soon to be an FOMC alum. Second,foracentralbanktoachievegood Nevertheless,beforeproceedingIwantto outcomesfortheeconomy,privatedecisionmakers emphasizethattheviewsIexpressherearemine needtounderstandmonetarypolicyandform anddonotnecessarilyreflectofficialpositions sensibleexpectationsaboutfuturepolicyactions. oftheFederalReserveSystem.Ithankmycol- Almosteverydecisionfirmsandhouseholds leaguesattheFederalReserveBankofSt.Louis makedependsinsomewayonexpectationsabout fortheircomments,butIretainfullresponsibility thefuture.Inflationexpectationsareofparticular forerrors. concerntoeverycentralbank—oreveryrespon- Itisnaturalformetobeinareflectivemood. siblecentralbank,anyway.Forfirmsandhouse- My10yearsattheSt.LouisFedhavebeenan holds,whatgoodstobuyandwhentobuythem, incredibleexperience.Iamsoontobecomean thewagestopayandtoaccept,andthefinancial outsideFedwatcher,asIwasbeforecomingto commitmentstomakedependonexpectations St.Louis,butmyperspectivewillbedifferent. aboutthepricelevelinthefuture.Ideally,wetake That,inpart,willbethesubjectofmyremarks forgrantedthestabilityofthepurchasingpower thisafternoon. ofmoneyanddonotconsciouslyconsiderinfla- tionexpectationsinourdecisions.Theidealstate oftakingforgrantedthepurchasingpowerof THE STRATEGY OF MONETARY moneyisincrediblyimportantforamarketsys- tem,andmanythingschangewhenwestartto POLICY incorporateconsciouslylikelyinflationinevery Oneofthekeylessonsoftherationalexpec- decision.Equallyimportant,oureverydaydeci- tationsrevolutioninmacroeconomicswasthat sionsdependonexpectationsaboutthestateof wemustthinkofeverypolicyactionasfitting therealeconomy—whethertheeconomywillbe withinapolicystrategy.I’llnowdroptheword saggingorsurging. “strategy”toemphasizethatamonetarypolicy Outcomesforinflationandtherealeconomy isthegeneralmodeofcentralbehaviorthatdeter- dependinpartonwhatthecentralbankdoes, minesindividualpolicyactions,orsettingsof whichinturndependsoncentralbankobjectives. thepolicyinstruments.Weneedtothinkinterms Allthisisfamiliargroundfromworkinmacro- ofapolicyfortworeasons.First,thecentral economicsoverthepast50years,butIrepeatit bank—andeveryotherdecisionmakingbody, sothatIcantellacoherentstory. publicorprivate—shouldnotactonthebasisof Muchofmythinkingoverthepast10years whim.Theremustbesomesystemofaction,for, hasbeendevotedtothissubject.Whatistheright otherwise,howcouldwepossiblyknowwhat policy?Thatis,howshouldindividualpolicy 1 MISCELLANEOUS actionsbefitintoageneralpolicyandnotbe,or makingthosedecisions.Iconfessthat,initially, appeartobe,drawnatrandom?Ihavegivena thisresponsewassomethingofadodge,because numberofspeechesonthistheme.Weclearly Iusuallyhadaprettygoodideaweeksinadvance havemadeprogressinthinkingaboutpolicy ofwhatmyownpositionatameetingwouldbe. actionsinthisgeneralway,althoughthereisa However,overtheyearsIhavebecomeimpressed longwaytogotomakethepolicyreactionfunc- byhowoftenmyownpositionwouldchangeeven tionmoreprecisebothtoguidepolicyactions inthedaysjustbeforeameetingasaconsequence themselvesandtomakethoseactionsmorepre- ofthearrivalofnewinformation,includingstaff dictabletothemarkets. analysisandsoundargumentsbymyFOMCcol- leagues.Itisnotthatmyviewsarepushedthis wayandthatbyarrivalofthelatesteconomic CENTRAL BANK datareports.Whathappensisthat,fromtimeto COMMUNICATIONS time,compellingnewinformationdoesarrive.I hopethatmypolicyoutlookwasstableevenas WhenIcametotheSt.LouisFed,Iwaswell- myviewontheappropriatepolicyactionmight preparedformyFOMCresponsibilitiesinmost changeinthelightofincomingdata. respects.Iknewalotaboutmonetaryeconomics Thinkingthroughthematterledmetoabitof andmonetaryhistory.WhatIdidnotknowwas research.WorkingwithBobRasche,theSt.Louis theartofcommunicatingwiththepressand Fedresearchdirector,Ihadalreadystudiedthe generalpublic.Theprofessionalliteraturein accuracyofmarketexpectationsaboutFOMC economicswasfullofinsightsintotheimpor- decisionsusingdatafromthefederalfundsfutures tanceofprivate-sectorexpectationsaboutmone- marketthedaybeforeeachFOMCmeeting.Given tarypolicybutessentiallysilentonhowthose thatthosefuturesmarketforecastshaveproven expectationswereformed,exceptfortheassump- tobequiteaccurate,anobviousquestionwas tionthatexpectationswouldnotbesystematically forecastaccuracylongerinadvance.BobandI wrongandwouldconvergetobeingcorrecteven- studiedfuturesmarketpredictionsofthefedfunds tually.OnceIstartedfieldingquestionsfromthe ratethreeandsixmonthsinadvanceandfound pressaftermyspeechesandtalkinginformally thattheaccuracywasprettylow.Thereason beforeawiderangeofaudiences,Iwaspartofthe theseforecastshavenotbeenverygoodisthat processoftryingtoestablishcorrectexpectations. newinformationarrivesthatcallsforachanged Mygeneralapproachhasbeentospeakpri- marilyaboutthepolicyprocessratherthanthe expectationonthemonetarypolicysetting,both specificsituationfacingtheFOMCatitsnext forthemarketsandfortheFOMC.Inotonly meeting.Itrytothinkofmyselfasspeakingto becamemoreawareoftheneedformetokeep portfoliomanagerswhohaveamedium-term mymindopenbutalsothoughtitimportantto horizonratherthantotraderswhohaveahorizon explaintomyaudienceshowthepolicyprocess measuredinhoursorafewdays.Idonotdispar- workedtoberesponsivetonewinformation. agetraders—theyperformanimportantfunction. Duringthepast10years,theFOMChasoften Obviously,Ihavehadinternalinformationthat discussedthenatureofthepolicystatementtobe wouldbeofinteresttotradersbutitwouldbe issuedattheconclusionofeachpolicymeeting. entirelyinappropriate—indeedillegal—todis- Arecurrentissueistheextentofforwardguidance closeconfidentialFOMCinformation. incorporatedinthestatement.WhenIjoinedthe Traders,portfoliomanagersandmanyothers FOMCinMarch1998,theCommittee’spractice alwayswanttoknowmyforecastofwhatwill wastoreleaseastatementonlyfollowingameet- happenattheupcomingFOMCmeeting.Mystan- ingatwhichitchangedthefederalfundsrate dardansweristhatIdonotforecastmonetary target.TheCommitteereleasedmeetingminutes policydecisions—myjobistoparticipatein shortlyafterthefollowingmeeting. 2 Reflections Forsomeyears,theCommittee’spracticehad mitteeproceededtoreducethetargetfedfunds beentoadoptastatementtobeincludedinthe ratebyatotalof75basispointsinthree25-basis- minutesaboutsymmetryinthedirective.The pointssteps.Thepreviouslystatedbiastoward directiveissuedtotheOpenMarketDeskmight tighteningdidnotcomplicatetheCommittee’s besymmetrical,orhaveabiastowardeithereas- actiontodealwithfinancialturmoil;Irecount ingortighteningpolicy.By1998,thebiashadno theepisodetoillustratehowquicklyconditions practicaleffectonDeskoperationsbutdidindi- canchange,overturninganearlierforecastofthe likelydirectionofthetargetfedfundsrate. catetothosewhoreadtheminutesthattheCom- TheFOMCcontinuedtodiscussthemerits mitteemightbemoreinclinedinthefutureto ofprovidingformalguidancetothemarketon raisethefundsratetargetthanlowerit,orvice thelikelydirectionoffuturepolicy.Atits versa. December1998meeting,theCommitteedecided TheminutesofthemeetingofMay19,1998, toreleaseastatementwheneveritsviewofthe releasedinearlyJuly,containedthispassage, balanceofriskschangedmaterially.1Thefirst whichwastypicalofthesymmetrylanguagein suchstatementcameinMay1999,whenthe 1998andearlieryears:“Allthememberswho Committeeleftthefundsratetargetunchanged intendedtovoteforanunchangedpolicyatthis butindicatedabiastowardraisingthetargetin meetingsupportedtheretentionofadirective thefuture.Themarketdidreacttothatstatement— thatwasbiasedtowardrestraint.Intheirview, theCommitteethoughtthemarketover-reacted. currentdevelopmentsdidnotcallforanypolicy TheCommitteeraisedthetargetby25basis action,atleastatthismeeting,butbecausethey pointsatitsJune1999meetingandbyanother felttherisksweretiltedinthedirectionofrising 25basispointsatitsAugustmeetingbutdidnot inflation,apolicytighteningmove,possiblyin expressadirectionalbiasineitherofthestate- thenearfuture,wasalikelythoughnotan mentsaccompanyingthesepolicyactions. inevitableprospect.”TwoFOMCmembers,both Atthetime,IrememberthatIwasverymuch favoringanincreaseinthefederalfundstarget infavoroftheFOMCprovidingwhateverguid- rate,dissented. anceitcouldonitsthinkingaboutfutureadjust- AlthoughtheFOMCwasapparentlyonthe mentsinthefedfundsratetarget.IftheCommittee edgeofraisingthetargetfedfundsrateinthe wereontheedgeofraisingthetarget,forexample, springandsummerof1998,theenvironment whynotrevealthatinformation?Thenthemarket thenchangedrapidlybecauseofthefinancial couldobserveincominginformationanddecide marketturmoilcreatedbyLong-TermCapital whetherstrongdatareportsmightcompletethe Management.ThefocusoftheCommitteechanged caseforraisingthetarget.Thatwayofthinking fromapredominantconcernovertheriskofrising aboutthematterwouldbesimilartothewayI inflationtothefinancialturmoil,andtheCom- thoughtaboutit. 1 ThispassagewasintheminutesoftheFOMCmeetingofDecember22,1998:“DisclosurePolicy.Themembersalsodiscussedvariousissues relatingtothetimingandmannerofreleasinginformationabouttheCommittee’spolicydecisions.Arangeofviewswasexpressed,asat earliermeetings,onthedesirabilityofreleasingastatementroutinelynotonlyafterthosemeetingsatwhichtherewasachangeinthestance ofpolicybutalsoaftermeetingswheretheCommitteealtereditsviewofthedirectionofpossiblepolicyactionsduringtheintermeeting period.Memberswhofavoredmoreannouncementsbelievedthatsuchdisclosure,byprovidingmoreinformationontheCommittee’sviews oftherisksintheeconomicoutlook,generallywouldallowfinancialmarketpricestoreflectmoreaccuratelythelikelyfuturestanceofmone- tarypolicy.However,othermemberswereconcernedthatsuchannouncementsoftenwouldprovokemarketreactions.Asaconsequence, theCommitteewouldbecomelesswillingtochangethesymmetryinthedirective,andapolicyofimmediatereleasemightthereforehave adverserepercussionsontheCommittee’sdecision-making.Nonetheless,themembersdecidedtoimplementthepreviouslystatedpolicyof releasing,onaninfrequentbasis,anannouncementimmediatelyaftercertainFOMCmeetingswhenthestanceofmonetarypolicyremained unchanged.Specifically,theCommitteewoulddosoonthoseoccasionswhenitwantedtocommunicatetothepublicamajorshiftinits viewsaboutthebalanceofrisksorthelikelydirectionoffuturepolicy.Suchannouncementswouldnotbemadeaftereverychangeinthe symmetryofthedirective,butonlywhenitseemedimportantforthepublictobeawareofanimportantshiftinthemembers’views.Onthe basisofexperiencewithsuchannouncements,theCommitteewouldevaluatelaterwhetherfurtherchangesinitsapproachtodisclosures wouldbedesirable.” 3 MISCELLANEOUS Thelanguagetoexpressthisideawaschanged thefundsratetargetunchanged,afterincreasing frompolicy“symmetry,”whichcouldincludea itby25basispointsateachofitsprevious17 biastowardraisingorreducingthefundsrate meetings.However,thestatementindicateda target,tothebalance-of-riskslanguageusedmore biastowardafurtherincreasebysayingthis: recentlybutexpressingmuchthesameidea. “Nonetheless,theCommitteejudgesthatsome Severalproblemsarose.Onewasthatitwashard inflationrisksremain.Theextentandtimingof forthemarkettodeterminewhenincomingdata anyadditionalfirmingthatmaybeneededto madethecasetochangethefundsratetargetas addresstheseriskswilldependontheevolution suggestedinthebalance-of-risksstatement. oftheoutlookforbothinflationandeconomic Anotherwasthatsometimesthedatacameinthe growth,asimpliedbyincominginformation.” otherway,inwhichcasetheCommitteemight JustaheadoftheAugustmeeting,themarkethad wanttochangethetargetintheoppositedirec- assignedaprobabilityofabout0.8onanFOMC tiontothatsuggestedearlier.Inprinciple,there targetfedfundsrateof5.25percentandaproba- shouldbenoproblemherebecausetheCommit- bilityofabout0.2onatargetrateof5.5percent.3 teemadeclearthatthepolicystancewasalways JustaftertheAugust2006FOMCmeeting,the contingentonincominginformation.Still,itjust marketassignedtheseprobabilitiestotheFOMC seemedawkward,tomeatleast,tochangethe decisionatitsforthcomingSeptembermeeting:a fundsratetargetintheoppositedirectiontothat targetof5.25percenthadaprobabilityofabout indicatedonlyashorttimebefore. 0.78,atargetof5.5percenthadaprobabilityof Ialsobecametroubledbythefollowingargu- about0.2percent,andatargetof5.75percenthad ment.Ifcurrenteconomicconditionsweresuch aprobabilityofabout0.02percent.Althoughthe tosuggestahighprobabilitythatfutureeconomic FOMCretainedthelanguagethat“firmingmay conditionswouldjustifyafutureincreaseinthe beneeded”atsubsequentmeetings,overtimethe fundsratetarget,whynotjustraisetherateat marketlowereditsprobabilitythattheFOMC thecurrentmeeting?Givenlagsintheeffectsof wouldinfactraisethetargetrate.Aheadofthe policyactions,thecurrentpolicyhadtobebased onthefutureoutlook.Ontheotherhand,ifthe FOMCmeetingofJanuary30-31,2007,themarket probabilitywerelow,woulditservethecauseof placedessentiallyzeroprobabilityonanytarget goodcommunicationtostateabias?Wouldn’tit rateabovetheprevailingrateof5.25percent. bemorehelpfultoworkhardertoarticulatethe AtitsmeetingofMarch20-21,2007,the conditionsunderwhichtheCommitteemight Committeedroppedthelanguagereferringto changethetarget—toexplaininmoredetailthe possiblefirming.Doingsomadelittledifference natureofthepolicyruleorresponsefunction?2 giventhatthemarkethaddiscountedthepossi- Anotherproblemwithforwardpolicyguid- bilityoffirmingforsometime. ancewasthataslowaccumulationofinformation Ihaverecountedseveraloftheseepisodesin sometimesmadethepriorbalance-of-riskslan- somedetailtoillustratethegeneralissue.Asa guageoutofdate,butitwasnoteasytotakeitout consequenceofobservingthisprocessfor10 ofthestatementwithoutsendingamessage,or years,IhaveconcludedthatanFOMCattemptto seemingto,thatafuturepolicyadjustmentinthe provideforwardguidanceinthepolicystatement otherdirectionwascontemplated.Thisproblem causesmorecommunicationsdifficultiesthanit arosein2006.InAugust2006,theCommitteekept solves.Akeyreasonisthattheeconomyissub- 2 Theseconsiderationsandothersledtotwospeeches,“MonetaryPolicyRules?”(FederalReserveBankofSt.LouisReview,March/April1999, 81(2))and“TheFed’sMonetaryPolicyRule”(FederalReserveBankofSt.LouisReview,January/February2006,88(1)). 3 IamrelyingontheprobabilityestimatesreportedbytheFederalReserveBankofClevelandderivedfromtradinginoptionsonfederalfunds futures[http://www.clevelandfed.org/research/policy/fedfunds/archives.cfm].IhavereadtheprobabilitiesofftheClevelandFedcharts; theseapproximationsarecloseenoughformypresentdiscussion. 4 Reflections jecttomoreshocksandreversalsthanonemight SHOCKS AND MORE SHOCKS think.Theseshockssometimesrequiremorefre- Tenyearsago,mybaselineapproachwasto quentpolicyactionsthanIwouldhavethought thinkoftheeconomyasgrowingalongitsfull- likelywhenIcametoSt.Louis.Ataminimum, employmentgrowthpath,orconvergingtothat changingeconomicconditionschangethelikeli- path,exceptwheninterruptedbyoccasional hoodthattheFOMCwillwanttoadjustthefed shocks.Theseshocksmighthavenegativeimpli- fundstargetinthedirectionpreviouslythought. cations,like9/11,orpositiveoneslikethesurge DirectionallanguagetendstoremainintheFOMC inproductivitygrowthstartinginabout1995.My policystatementbeyondthetimeitappliesand viewaboutshockshaschanged:theshocksare removingthelanguagecreatesthepossibilityof continualandnotoccasional.Overthepast10 miscommunication.Everychangeinthepolicy years,therehavebeenmanymoremonthsin statementleadsnaturallytomarketquestionsas whichtheeconomywasbeinghitbysubstantial towhatthechangemeansandwhetherthechange surprises,oradjustingtothem,thantranquil ismeanttoprovideahintaboutthefuturedirec- months.Crudeoilpriceshavefluctuatedovera tionofpolicy.Tomymind,everytimenewlan- tremendousrange,ashaveequitypricesandnow guageisinsertedintothepolicystatement,there homeprices.Financialmarketswereshakenby needstobeasmuchthoughtgivenastohowto Long-TermCapitalManagement,Enronand exitfromthelanguageastotherationalefor WorldComandnowthesubprimemortgagemess. insertingit. TheterroristattacksofSeptember2001andthen Iknowthatmarketparticipantsarehungry theU.S.invasionofIraqin2003createduncer- forinsightintotheFOMC’sthinkingandintothe tainty.TheY2Keffortwassubstantial—inthe likelihoodoffutureadjustmentsinthetarget eventtherewasnoevent,butbeforetheevent federalfundsrate.Myjudgmentisthat,mostof therewasuncertaintyandrisk. So,surprisesarenormalandtheirimplica- thetime,theCommitteecannotprovidewhat tionsforcentralbankpracticeareoften—indeed themarketwantsbecausetheCommitteeitselfis ordinarily—unclearaheadoftime.Itrytosepa- notclairvoyant.Nooneknowshowtheeconomy ratesurprisesrequiringcentralbankresponses isgoingtoevolveandhoweventswillchange fromthoseinwhicharesponsewouldbeunhelp- theappropriatesettingofthefederalfundstarget ful.Marketadjustmentsdodealwithmanysur- rate.Mostofthetimeoverthepast10yearsIhad prises—marketsaresuperbinthisrespect.Price hunchesaboutthepolicydirectionIwouldbe signalsalterdecisionsandequilibratechanges advocatingatthenextFOMCmeeting,but ingoodssuppliesanddemands. “hunches”reallyistherightword.Ihadhunches TheabstractframeworkIsketchedattheout- andnotsettledconvictions.Furthermore,the setprovidesconsiderableguidance.Thequestion moreIreflectedandthemoreexperienceIaccu- isalwayswhetherapolicyresponsetoashock mulated,themoreIrealizedhowfrequentlysur- willhelpinachievingthegoalsofpricelevel prisechangesinconditionsrequiredthatIchange stabilityandarealeconomyclosetoitslong-run myhunches.Ishouldnotbemisinterpretedas growthpathdeterminedbyproductivityandthrift. sayingthatInecessarilychangedmyviewonthe Everypolicyactionshouldbeviewedasbeing appropriatesettingofthefedfundsratetarget. logicallyderivedfromthegeneralpolicyandthe Butwhentheinformationonwhichmyprior currentfacts.Ifapolicyresponseitselfwouldbe hunchwasbasedchangedsignificantly,Ihadto ashock,becauseitcouldnotbeanticipatedfrom startover,inasense,tofigureoutwhetherthe thegeneralpolicyrule,thenthatcreatesinmy newinformationrequiredachangeinthepolicy mindapresumptionagainsttheproposedpolicy stance. action. 5 MISCELLANEOUS OTHER MATTERS AlthoughtheFederalReservehasacultureof veryhighintegrity,reinforcingthatcultureis Ihaveconcentratedmyremarksonmonetary obviouslyworthwhile.TheproblemtheFederal policyissuesbecausethatisthesubjectthathas Reservefaces,incommonwithothernot-for- takenmostofmytimeoverthepast10years.But profitfirms,isthatemployeesdonotalwayspush therehasbeenalotmoretomySt.LouisFedjob hardenoughtogetthingsdone.Wetendtosuffer thanmonetarypolicy. fromparalysisbyanalysisandexcessivelyhigh IknewlittleaboutmanagementwhenIcame risk-aversion.MysenseisthattheSt.LouisFed toSt.Louis,atleastinanyformalsense.Never- hasmovedintherightdirectioninrecentyears. theless,themanagementsideofbeingaCEOhas Thereisalotmoreinnovativeactivityatthe turnedouttobeveryinterestingtome.Ihave Bankthanoutsiderscanobserve. learnedalotaboutgoodmanagementpractice, includingtheroleofastrongboardofdirectors, internalauditandothercontrolprocesses,and CONCLUDING COMMENT thetaskofstayingintouchwithemployeesin allpartsoftheBankatalllevels.TheSt.Louis AllaspectsofmytimeattheSt.LouisFed Fedhasroughly1,000employees.Ifinditmind- havebeeninterestingandrewarding.Ihavenot bogglingthatsuccessfulCEOsofmuchlarger hadanythingtodoasburdensomeasmyexam- organizationsareabletostayintouchwithall gradingchoreswhenIwasanacademic. thefirm’semployees.I’veusedpersonalappear- IleavetheFedattheendofnextmonthwith ancesbeforelargegroupsofemployees,internal adeepappreciationofthecompetenceandenergy mediaandselectivemeetingswithsmallgroups levelofFedemployees.TheFedknowswhatit ofemployees.Althoughitisimpossibletomeet isdoing,uptothelimitsimposedbythestateof allemployeesinsmallgroups,wordspreadsfrom knowledgeineconomics.Professionalstandards themeetingsthatdotakeplace.Atleastthathas aretheveryhighest.Peopleworkhardandnever beenmystrategy. letpoliticsinterferewithdecisions.Simplyput, Effortstostayintouchwithemployeesare thishasbeenafabulousexperienceforme. importanttoprovideclearexpectations,motiva- tionandpromoteasoundcorporateculture. 6

Cite this document

APA

William Poole (2008, February 10). Speech. Speeches, Federal Reserve. https://whenthefedspeaks.com/doc/speech_20080211_poole

BibTeX

@misc{wtfs_speech_20080211_poole,
  author = {William Poole},
  title = {Speech},
  year = {2008},
  month = {Feb},
  howpublished = {Speeches, Federal Reserve},
  url = {https://whenthefedspeaks.com/doc/speech_20080211_poole},
  note = {Retrieved via When the Fed Speaks corpus}
}