speeches · May 17, 2006

Speech

William Poole · President

Inversion GlobalInterdependenceCenter CentralBankSeries FederalReserveBankofPhiladelphia Philadelphia,Pennsylvania May18,2006 Inversion has been much in the news for andthe10-yearrate,butlookingforsomepartic- some months now. Indeed, inversion has ularpartoftheyieldcurvewhereashorterrate made the big time, with William Safire isabovealongerrateisscratchingforastory. devoting a column in the Sunday New Surely,noimportantissuecandependonan York Times (April 23, 2006) to IYC—inverted inversionsomewherealongtheyieldcurveofa yield curve. By inversion, of course, I’m refer- fewbasispoints. ring to a situation in which short-term interest Beforeproceeding,Iwanttoemphasizethat rates are higher than long-term interest rates. theviewsIexpressherearemineanddonot WhenIagreedtospeakonthistopiclastfall, necessarilyreflectofficialpositionsoftheFederal marketconcernoverIYCwasrunninghigh. ReserveSystem.Ithankmycolleaguesatthe TheFOMChadbeenprovidingguidancethatit FederalReserveBankofSt.Louis—especially wouldprobablycontinuetoraisethetargetfederal MichaelDueker—fortheirassistanceandcom- fundsrate;giventhelevelofthe10-yearTreasury ments,butIretainfullresponsibilityforerrors. yieldinthe4¼to4½percentrange,market observersexpectedthatthefundsratewouldsoon beabovethe10-yearrate.Recessionconcerns TERM-STRUCTURE THEORY werewidelydiscussed,becauseinthepastIYC AnimportantproblemwithmuchIYCcom- hasoftenbeenassociatedwithrecession.More- mentaryisthatitinvolvesasearchforpatterns over,manyfounditoddthatuntillastmonththe inthedatawithoutanefforttounderstandthe monthlyaverage10-yearbondratewasactually lowerthanithadbeeninJune2004,whenthe economicsthatmightliebehindthepatterns. FOMCbegantoraisethefedfundsratetarget.It Understandingwhyobservationsfollowaparticu- seemedapuzzlethatthe10-yearratewasactually larpatternisessentialtojudgingwhetherapattern thesameinMarch2006asithadbeeninJune islikelytopersistorapplytotoday’ssituation. 2004,eventhoughtheFOMChadraisedthetarget Economistshavebeenstudyingtheterm fedfundsratefrom1percentto4¾percent.Now structureofinterestratesforalongtime.The thatthe10-yearratehasrisenbyanother50to firstpropositionisthatalonginterestratereflects 75basispoints,toabout5.15percent,apparently investorexpectationsoftheaverageshortrate everyonefeelsalotbetter! overthehorizonofthelongrate.Thus,today’s Forsimplicity,usingmonthlyaveragedata, 1-yearratereflectsexpectationsofthenext521- exceptwhereindicatedotherwise,I’llconcentrate weekrates;today’s10-yearratereflectsexpecta- onthedifferencebetweenthe10-yearconstant- tionsofthenext101-yearrates. maturityTreasuryyieldandthefederalfunds Ideliberatelyusedthephrase“reflects rate.Someanalyststhispastwintercalledatten- expectations”becausethereisampleevidence tiontoaninversionbetweenthe1-or2-yearrate thatalongrateisnotalwaysequaltotheappro- 1 FINANCIALMARKETS priatelyweightedaverageoftheshortratesover atedwiththerecessionsof1957-58and1960-61. thehorizonofthelongrate.Mostofthetime,long Beforethemid-1960s,theFedadjustedRegulation ratesaresomewhataboveshortrates.Overthe Qinterestceilingsinafashiontimelyenoughto past50years,forexample,the10-yearTreasury preventsignificantdisintermediation. ratehasaveragedabout90basispointsabovethe InanenvironmentofslowadjustmentofReg federalfundsrate.Thedifferencebetweenthe Qceilings,whentheFedsteppedonthemonetary longrateandaverageexpectedshortrateover brakes,bankcreditbecametightandthereal thehorizonofthelongsecurityiscalledthe short-terminterestratequicklyrosewellabove “termpremium.” itsequilibriumlevel.Becausethemarketantici- Onaverage,thetermpremiumispositive, patedthatthemonetarybrakeswouldbeloosened buttheorydoesnotpredictanyparticularrela- withinarelativelyshorttimeframe,long-term tionship.Thetermpremiumisthoughttoarise interestratesrosebyamuchsmalleramount, frominvestorattitudestowardrisk.Capitalvalues andtheyieldcurveinvertedtemporarily.Itwas fluctuatemorethelongerabond’smaturity,and duringthiserathatyield-curveinversionscame investorsaversetocapitalriskthereforeprefer tocarrynegativebusiness-cycleconnotations. shortermaturities.Ontheotherhand,interest Alltoooften,theclampdownoncreditwas incomefluctuatesmoreforaseriesofinvestments severeenoughtobeassociatedwitharecession inshort-termbondsthanforalong-termbond; butnotsteadfastenoughtobringaboutlasting investorsaversetoincomeinstabilitytherefore disinflation. preferlonger-termbonds. Fortunately,sincetheearly1980s,three Thebalanceofinvestorswithdifferentand thingshavechangedforthebetter.First,lasting changingattitudestowardriskchangesovertime, disinflationwasachievedintheVolcker- andotherconditionsmayalsochange.Thus, Greenspanera.Second,financialderegulation thereisnoreasontoexpectthattermpremiums didawaywithRegulationQ,andimportantfinan- willbeconstant,andtheyaren’t.Giventhat cialinnovations,suchashedginginstruments, investorexpectationsaboutfutureshortratesare havelessenedtheeconomicfalloutfromshocks. notdirectlyobservable,andtheirpreferences Third,theeconomyitselfhasbecomemoresta- thatcreatetermpremiumsarenotdirectlyobserv- ble—aphenomenonknownasthe“Great ableeither,thereisnoabsolutelyreliablewayto Moderation.”Moreover,thesepositivedevelop- disentanglechanginginterestrateexpectations mentsreinforceeachotherbecause,asChairman fromchangingtermpremiums. BernankehasnotedinCongressionaltestimony, amorestableenvironmentprovidesmonetary policymakersgreaterlatitudetorespondaggres- SOME HISTORY sivelytoshocks.1Itwasthislatitudethatpermit- tedtheFederalReservetomaintainthefederal Manyoftheinversionsoftheyieldcurve fundsrateat1percentfromJune2003toJune startinginthe1960soccurredundertheold 2004whilewaitingpatientlyforthecurrent depositinterestrateceilingsestablishedunder expansiontoestablishfirmlegs. RegulationQ.Theceilingonthedepositrateled to“disintermediation”fromthebankingsystem whenmonetarypolicytightenedandincreased IYC AS RECESSION PREDICTOR theresponsivenessofthequantityofmoneyinside thebankingsystemtoFederalReservepolicy Onequipaboutthepredictivepowerofyield actions.However,therewasnoinversionassoci- curveinversionsisthattheyhavepredictedsix 1 TestimonyofChairmanBenS.Bernanke,SemiannualMonetaryPolicyReporttotheCongressbeforetheCommitteeonFinancialServices, U.S.HouseofRepresentatives,February15,2006. 2 Inversion ofthelastfourrecessions.Infact,startinginthe of1974,andtheFedeasedpolicy.Theinversion mid-1960s,therehavebeenninenotableinver- lastedthroughDecember. sionsoftheyieldcurveandsixbusinessreces- ThestoryaroundthecyclepeakinMarch sions.Thehistoricalrecordofyieldcurve 2001wasabitdifferent.TheFedmaintaineda inversions,recessions,andfalsealarmsreveals targetforthefundsrateof6.5percentstartingin severalregularities.Suchananalysisalsosheds mid-May2000.AfterMay,the10-yearTreasury lightontoday’srelativelyflatyieldcurve. ratefellabitandtheyieldcurveinverted.The Thebasicpropositionisthattheyieldcurve inversionreachedamaximumof116basispoints reflectsinvestorexpectationsoffutureinterest inDecember.Policyin2000wasdominatedby rates.Theseexpectationsdependonanticipated concernoverthethreatofinflationratherthan Fedpolicyadjustments,whichinturnreflect actualinflation.TheFOMCbegantocutthefunds expecteddevelopmentsintherealeconomyand ratetargetinJanuary2001,aheadofthecycle therateofinflation.Consideraclassicepisode peakinMarch.TheinversionendedinAprilas toillustratetheprocess.Inflationbegantorise theFedcutthefundsratetargetaggressively. persistentlyin1967and,afterthemiddleofthe Whileitistruethataninversionprecededthe year,theFedbegantotightenpolicy.Inresponse cyclepeak,itisalsotruethattheFOMCbeganto totighterfiscalpolicyenactedinmid-1968,which cutratesbeforethecyclepeakandindeedcut manybelievedwouldcooltheeconomy,theFed aggressivelybecauseinflationwascontrolled. easedpolicyabit.However,theeconomydid NowthatI’vereviewedafewepisodes,let’s notslowandinflationcontinuedtoworsen.The trytogeneralizeabit.Usingweeklyaveragedata FOMCthenraisedthefedfundsratealmostevery totiedowntimingrelationshipsaccurately,the month.Theyieldcurvehadinvertedslightlyin lastdateonwhichtheyieldcurveisinvertedby thespringof1968,andtheinversiondeepened atleast100basispointscanbeconsideredthe astheFedtightenedpolicy.InAugust1969,the dateonwhichtheinversionebbsandstartsto peakmonthforthefundsrate,theinversion tailoff.We’llcallthisdatethe“ebbdate.”One reacheditsmaximumextentof250basispoints. interestingpatternacrossyieldcurveinversions ThebusinesscyclepeakwasDecember.Notuntil isthatthelongratetypicallyrises,althoughsome- March1970didtheFedeasepolicyenoughto timesonlyslightly,inthethreemonthspriorto bringthefundsratedownsignificantly. thestartoftheinversionandthenfallsinthe Thebasicstory,asIseeit,wasthattheFed threemonthspriortotheebbdate.Itisnotobvi- wasslowtoeasepolicyin1969and1970because ousthatthelongratewouldbefallingasthe itwasconcernedabouttheinflationrate.The inversionisdeclining,butthispatternheldin bondmarketcouldseethattheeconomywas manyoftheinversionsthatprecededorcoincided weakeningandthatrateswouldbecomingdown, withrecessions—specificallytheinversionsthat whichiswhytheinversiondeveloped.Butthe startedin1968,1973,1978,1982,1989,and2000. Feddidnotwanttoeasepolicyuntilinflation Thesecasesfitthescenariowherebymonetary slowed,bothbecauseitwantedtomaintaindown- policyinitiallytightensenoughtoliftallinterest wardpressureontheinflationrateandbecause rates,includinglong-termrates,butaweakening itwasconcernedthatprematureeasingwould economyandmarketanticipationsofFedeasing raiseinflationexpectations.Thissamepattern subsequentlyleadtodeclinesinlong-termrates. playedoutinspadesinthemonthsaroundthe TheinversionebbsbecausetheFedseesthe November1973businesscyclepeak.Inflation weakeningeconomyandbringsthefundsrate wasmuchhigherandinflationexpectationsmore down,andthefundsratedeclinesmorequickly entrenched.Usingmonthlyaveragedata,thepeak thandoesthelongrate. fedfundsratewasinJuly1974andtheinversion Itisalsoworthnotingthatinallsixofthese thatmonthwasover500basispoints.Theecon- classicinstanceswhereaninversionoftheyield omyweakeneddramaticallyinthesecondhalf curveprecededarecession,therealfederalfunds 3 FINANCIALMARKETS rate—measuredasthedifferencebetweenthe short-terminterestratewhengaugingwhether federalfundsrateandcorePCEinflation— recessionislikely. exceeded4percentandsometimesbyawide Asarecessionpredictor,yield-curveinver- margin.Thesewereepisodesofsubstantialpolicy sionsdonotoutperformothersimplerulesof restraint,motivatedbyinflationorthethreatof thumb,suchastroughsintheunemployment inflation. rate.Eventhoughtheunemploymentrateis Therearethreefalsealarmcaseswhereno widelyknownasalaggingeconomicindicator,a recessionensued—theinversionsthatbeganin simplepredictiveruleisthatarisefromatrough 1966,1995,and1998.Innoneofthesecasesdid levelintheunemploymentratebyatleast0.3 theinversionreach100basispointsusingmonthly percentagepointslastingatleastthreemonths averagedata.Inthefalsealarmcases,thereal occurspriortoeverycyclepeak.Forexample, federalfundsratewasatorsomewhatbelow3 byAugust2000,theunemploymentratehad percent. risen0.3percentagepointsaboveanApril2000 Theanomalyamongtheinversions,inthatit troughrateof3.8percent.Therecessionthen isneitheroneoftheclassiccasesnorafalsealarm, begansevenmonthslaterinMarch2001.Never- istheinversionthatstartedinOctober1980.The theless,theeightcorrectsignalsfromtheunem- ebbdateofthisinversion,inSeptember1981, ploymentrateareaccompaniedbythreefalse occurredinratherunusualcircumstances.The alarms—withthelatestinFebruary1986.This economyhadgonebackintorecessioninJuly ratioofhitstomissesissimilartotheratiofor 1981andthe10-yearratewasnearitspost-war yieldcurveinversions.Undoubtedly,wecould highofabout15½percentattheheightofthe findmanyotherrecessionsignalsthatwould Fed’sstruggletobringinflationdown.Although matchtherecordofinvertedyieldcurves.That bySeptember1981thefedfundsratehadcome allthesesupposedsignalsareoflimitedvalueis downfromitspeakearlierintheyear,thepolicy indicatedbythefactthatforecastersdonothave outlookwasextremelyuncertainbecausethe astellarrecordofforecastingrecessions.Ifthe inflationoutlookwassouncertain.Asitturned signalswereclear,forecastingwouldbeeasy.It out,theeconomyweakenedrapidlyasthereces- isnot. siontookholdandbothlongandshortrates Ihaven’tattemptedacitationcounttodeter- declinedafterSeptember1981.Theinversion minewheninversionaversionreacheditspeak disappearedasshortratesdeclinedmorequickly intensity,butthediscussionwascertainlyactive thanlongrates,butthenreappearedforafew fromlatelastfalltoearlythisyear.Perhapswhat monthsin1982. triggeredthisdiscussionwasthat,whiletheFOMC Analyzingthecurrentsituationinlightof wasraisingthetargetfedfundsratestartingin thesepatterns,usingweeklyaveragedata,the June2004,the10-yearratetradedmostofthetime spreadbetweenthe10-yearTreasuryandthe inarangefrom4to4.5percent.Lastfall,asthe federalfundsrateneverbecamenegativethispast targetfundscontinuedtorisebutthe10-year winter,thoughitcameclosetozeroinJanuary ratedidnot,itappearedonlyamatteroftime andFebruary.Therealfederalfundsratehas untiltheinversionwouldoccur.Infact,using remainedbelow3percentinallof2006.Thus, monthlyaveragedata,wehavenothadaninver- therecentrelativelyflatyieldcurvehasmuch sionthroughApril2006. moreincommonwiththecaseswhereyieldcurve Thekeytounderstandingthissituationis inversionswerenotfollowedbyarecession. thatincreasesinthetargetfundsratewerewell Onelessonfromtheseepisodesisthatthe predictedinJune2004,whentheincreasesbegan yieldcurvemustbecombinedwithadditional fromtheunusuallylowfederalfundsrateof1per- informationinorderforareliablerecessionsignal cent.InJune2004,themarketcorrectlygauged toemerge.Inparticular,thetermspreadshould thattheFedwouldraisethefundsratesteadily beconsideredjointlywiththelevelofthereal andgraduallyforthenextyearandahalf.Not 4 Inversion untiltheNovember2005FOMCmeetingdidthe mostofwhatwe’veobserved.Butthereareno targetfundsrateexceedtheratethathadbeen doubtotherforcesatwork.Itappearsthatthe expectedinJune2004.Withthefundsraterising termpremiuminlongratesfellasthefundsrate ontheexpectedtrack,therewasnoreasonforthe targetincreased.Onelikelyreasonthattheterm 10-yearbondratetodepartinanymajorwayfrom premiumfellinthefirstyearandahalfofthis itslevelinJune2004.Theincreaseinthebond tighteningcycleisthatthemarketunderstood yieldsinceNovember2005isconsistentwiththe thepaththatshort-terminterestrateswouldtake ideathatthefundsratehasnowincreasedsome- inthetighteningcyclethatbeganinlateJune whatmorethanthemarketanticipatedearlier. 2004.Thatpredictabilityreducedtheriskof Whydidratesbehavesomuchthesameas holdinglonger-termbonds. expectedinJune2004?Themostimportantrea- Marketcommentaryhasattributedmuchof sonisthattheeconomycameinsoclosetowhat therecentincreaseinlongratestoarestoration hadbeenforecast.GoingasfarbackasJanuary ofamore-normaltermpremiumforholdinglong- 2004,theBlueChipforecastersmaintainedvery termdebt.Butpolicymakersshouldnotviewthe steadyandaccurateforecastsof2005GDPgrowth, termpremiumasasinglecomponentoflong-term whichcameinat3.5percent.Themainsurprise interestrates.Instead,thetermpremiumconsists wasintheinflationrate.Inturn,thatsurprise ofcompensationfortheriskthatrealinterestrates wasaconsequenceofenergyprices,whichhave willturnouttobehigherthanexpectedinthe increasedfarbeyondlevelspredictedinJune futureandseparatecompensationfortherisk 2004buteventodayhavenotmuchaffectedcore thatinflationwillturnouttobehigherthancur- inflation.Whathashappenedtoraisetheexpected rentlyexpected.Naturally,ifthetermpremium pathforthetargetfederalfundsratefromexpec- increasesbecauseofchangesinbearingreal tationsinplacelatelastyear,andthereforehas interestraterisk,asapolicymakerIammore alsoaffectedlonger-termrates,isthepersistence comfortablewiththatthanifthetermpremium ofenergypriceincreases.InJanuary2005,the increasesbecauseofmarketconcernsaboutthe BlueChipconsensusforCPIinflationfor2006 riskofinflation.YieldsonTreasuryinflation- was2.3percent.AsoftheMay2006BlueChip indexedsecurities,combinedwithfactormodels release,thatconsensusisnowupto3.1percent; ofthetermstructureofinterestrates,suggestthat muchofthisincreasetookplaceafterHurricane thecompensationforbearingrealinterest-risk Katrina. droppedbetweenJune2004andlate2005, Unlessweexperiencesignificantdepartures althoughithasreboundedsomewhatsincethen. fromtheexpectedcoursefortheeconomy,the Becausesomeofthesefactormodelssuggested recentriseinthe10-yearTreasuryyieldfrom thatthetermpremiumhaddippednearlytozero about4.6percentinFebruarytoabout5.15per- bylate2005,somereboundwaslikely. centtodayhasdonemuchtodiminishthelikeli- Amongtheinternationalfactorscitedas hoodofasubstantialinversionintheyieldcurve influencesonU.S.interestratesinthepastfew inthenearfuture. yearsistheglobalsavingglut.Unusuallyhigh savingmightholddownthelevelofrealinterest RECENT BEHAVIOR OF LONG- rates,butthereisnoreasonwhythereshouldbe aneffectontheshapeofthetermstructure.In TERM RATES AND THE TERM anyevent,itappearsthatrealinterestratesare PREMIUM returningtoamorenormallevelintheUnited I’veemphasizedtheimportanceofinterest States.The10-yearindexedbondhadarateof rateexpectationsforshapingtheyieldcurveand about1.6percentinthefallof2004;thatrateis believethattherateexpectationsstoryexplains nowuptoabout2.4percent. 5 FINANCIALMARKETS FORWARD RATES alreadynoted,thespreadbetweenthe10-year bondandthefedfundsneverbecamenegative Althoughtheyieldcurveisanimperfect lastfallandstillisn’t.Yet,inversionsassociated recessionpredictor,thetermstructureofinterest withrecessionshavebeenquitelarge.Using ratesprovidesveryusefulglimpsesofwhatshort- monthlyaveragedata,the1969inversionreached terminterestratesarelikelytobeinthefuture. 250basispoints;the1974inversionexceeded Aforwardinterestratefarenoughintothefuture, 500basispoints;the1980and1981inversions saynineyearsahead,providesinformation exceeded600basispoints.Milderinversions aboutthetrendrateofinflationmarketsexpect. seemtohavebeenassociatedwithmilderreces- Measuresofexpectedinflationintheshortrun, sions.The1989inversionreached125basis sayinthenexttwoyearsorso,reflectenergy pointsandthe2000inversionreached116basis priceshocks,forexample,thatwillnotinfluence points.Wenevergotclosetoanyoftheselast thelong-runtrendrateofinflation.Similarly,no fall.Finally,lookingbackat1980-82experience onecanforecastthestateofthebusinesscycle makesclearthatsimplycountingpresumedpat- nineyearsintothefuture,sotheimpliedfarfor- ternsinthedata,withoutguidancefromeco- wardratereflectsneutralbusinesscyclecondi- nomictheory,isnotawisestrategy.Theearly tions.Forthisreason,aclearshiftupordownof 1980sweresodifferentfromtoday’sconditions theimpliedfarforwardratesuggeststhateither insomanyrespectsthattheexperienceoftwin thetrendlevelofexpectedinflationhaschanged recessionsinahighinflationerahaslittlebear- orthemarket’sinferenceofthetrendrealrateof ingonunderstandingthetermstructuretoday. returnhasbeenaltered.Inprinciple,onecould Thetermstructureofinterestratesprovides usedataonstrippedinflation-indexedbondsto awindowintoinvestorinterestrateexpectations. infertheimpliedforwardrealrate,buttradingin Itisalwaysworthwhileforpolicymakerstocon- thismarketissomewhatthin.Nevertheless,the siderthoseexpectationsbutnotwisetotakethem impliedfarforwardrealrateisnotexpectedto atfacevaluewithoutfurtheranalysis.Interest- undergosuddenchanges.Hencewecaninterpret, rateexpectationsreflectinvestorunderstanding asagoodapproximation,changesintheimplied ofhowrateswillevolve,whichiswhyaninverted farforwardnominalrateasacombinationof yieldcurvehasoftenprecededbusinesscycle changesinlong-terminflationexpectationsand peaks.Butthemarket’srateexpectationsalso inthetermpremium. dependimportantlyonthemarket’sreadofwhat PricedatafromstrippedTreasurybondcoupons theFOMCwilldo.Ifthemarket’sexpectation doesnotmatchtheFOMC’sownexpectation,then showthattheimpliedfarforwardone-yearrate policymakersneedtodosomesoulsearching. nineyearsinthefuturehasfallensincetheFed Therearetwopossibilities:eitherthemarket initiateditstighteningcycleinJune2004.Inthat mayhaveabetterunderstandingofwherethe timespantheimpliedforwardratehasfallen FOMCislikelytotakepolicythantheFOMC fromabout61/ 8 percenttoabout5½percent.Thus, does,orthemarketmaybemisreadingtheFOMC’s thesignalfromtheTreasurybondmarketisthat intentions. theFed’smeasuredbutsteadypaceofremoving Arguably,themarketwasaheadoftheFOMC policyaccommodationhasbeensufficientto in2000;thepeakforthe10-yearratewas6.66 keeplong-terminflationexpectationsandrisk percentinJanuary,andbyDecembertherate premiawell-contained. wasdownto5.24percent.Asevidenceofthe economy’sweakeningaccumulated,theFOMC firstcutthefedfundsratetargetinJanuary2001. CONCLUDING COMMENTS Conversely,aftertheFOMCloweredthefunds ImustsaythatI’vebeenabitpuzzledbythe ratetargetto1percentinJune2003,themarket inversion/recessiontalkthatbeganlastfall.As seemedprimedtoexpectthattheCommittee 6 Inversion wouldsoonberaisingtherate.Tobetteralign marketexpectationswithitsown,initsAugust 2003meetingtheFOMCintroducedlanguagein itsstatementindicatingthat“theCommittee believesthatpolicyaccommodationcanbemain- tainedforaconsiderableperiod.” Atleastasoftoday,themarket’sconcerns lastfallthattheyieldcurvewouldinvertand signalarecessionseemtohaveevaporated.There isnoobviousmisalignmentofmarketinterest rateexpectationsandthelikelycourseofpolicy giveninformationavailabletoday.WhatIbelieve willhappenisthatFOMCpolicydecisionsand marketexpectationswillevolveasnewlyarriving dataeitherchange,oraffirm,thecurrentoutlook fortheeconomy.Thepolicystatementfollowing theFOMC’smeetingofMay10indicatedthe Committee’sviewthateconomicgrowthwason asolid,butmoderating,trackandthatinflation wascontained,butstillarisk. TheCommitteealsoemphasizedthatfuture policywoulddependonhowarrivinginforma- tionaffectedtheeconomicoutlook.Imyselfplace greatemphasisonthispoint.Experienceindicates thateconomicforecastsarenotespeciallyaccu- rate,andthatmeansthatmonetarypolicyactions shoulddependonhowtheoutlookchangeswith newinformationratherthanbedecidedin advancebasedontheforecast. IhopeI’veprovokedyousufficientlythat you’llasksomequestions.Fireaway. 7

Cite this document

APA

William Poole (2006, May 17). Speech. Speeches, Federal Reserve. https://whenthefedspeaks.com/doc/speech_20060518_poole

BibTeX

@misc{wtfs_speech_20060518_poole,
  author = {William Poole},
  title = {Speech},
  year = {2006},
  month = {May},
  howpublished = {Speeches, Federal Reserve},
  url = {https://whenthefedspeaks.com/doc/speech_20060518_poole},
  note = {Retrieved via When the Fed Speaks corpus}
}