Speech

The Fed’s Monetary Policy Rule CatoInstitute Washington,D.C. October14,2005 PublishedintheFederalReserveBankofSt.LouisReview,January/February2006,88(1),pp.1-11 In 1936, Henry Simons published a paper, policyhasbeenhighlypredictableoverthepast “Rules Versus Authorities in Monetary decadeorso.IfthemarketcanpredicttheFed’s Policy,” that not only became a classic but policyactions,thenitmustbethecasethatFed alsoisstillhighlyrelevanttotoday’spolicy policyfollowsarule,orpolicyregularity,ofsome debates.1 Irediscoveredseveralimportantpoints sort.Mypurposeistoexplorethenatureofthat in the paper while preparing this lecture. rule.ContrarytoSimons’simplication,thebehav- Inthinkingaboutpolicyrulesinrecentyears, iorofauthoritiescanbepredictable. Ihavetendedtoseparatethepoliticalandecon- Beforediggingintospecifics,considerwhat omiccasesforarule.Simonsarguesforamuch the“rulesversusdiscretion”debateisabout. moreintegratedviewoftheissue: Advocatesofdiscretion,asIinterpretthem,are primarilyarguingagainstaformalpolicyrule,and Thereare,ofcourse,manyspecialresponsi- bilitieswhichmaywiselybedelegatedto certainlyagainstalegislatedrule.Theybelieve administrativeauthoritieswithsubstantial thatpolicywillbemoreeffectiveifcharacterized discretionarypower...Theexpedientmustbe by“discretion.” invokedsparingly,however,ifdemocratic Discretionsurelycannotmeanthatpolicyis institutionsaretobepreserved;anditisutterly haphazard,capricious,random,orunpredictable. inappropriateinthemonetaryfield.Anenter- AdvocatesofdiscretionagreewithSimonsthat prisecannotfunctioneffectivelyinthefaceof “manyspecialresponsibilities...maywiselybe extremeuncertaintyastotheactionofmonetary delegatedtoadministrativeauthoritieswithsub- authoritiesor,forthatmatter,astomonetary stantialdiscretionarypower.”However,theydo legislation.2 notagreewithSimonsthatdiscretion“isutterly inappropriateinthemonetaryfield.” Thus,Simonsarguesthattheruleoflawthat Interestingly,Simonsarguedthatafixed characterizesademocracyisalsorequiredto providemonetarypolicypredictability,which, moneystockwouldbethebestrule,butonlyif inturn,isnecessaryforefficientoperationofa substantialinstitutionalreformswereinplacein marketeconomy. financialmarkets,suchas100percentreserve I’vechosenatitledesignedtobeprovocative, requirementsagainstbankdeposits.Giventhe forIsuspectthatfewconsidercurrentFederal institutionalstructure,Simonsarguedforarule Reservepolicyascharacterizedbyamonetary focusedonprice-levelstabilization,because“no rule.Mylogicisthis:Thereisnowalargebody monetarysystemcanfunctioneffectivelyorsur- ofevidence,whichI’llreviewshortly,thatFed vivepoliticallyinthefaceofextremealternations 1 Simons(1936). 2 Simons(1936,pp.1-2). 1 MONETARYPOLICYANDINFLATION ofhoardinganddishoarding.”3Thatis,Simons ments—especiallyBobRasche,seniorvicepresi- believedthatlargevariationsinthevelocityof dentanddirectorofresearch. moneywouldmakeafixedmoneystockrule workpoorly. POLICY PREDICTABILITY— Despitethenatureofhisargumentforaprice- levelstabilizationrule,elsewhereinthesame A SUMMARY OF FINDINGS paperSimonsarguedthat,“[o]ncewellestablished I’vediscussedthepredictabilityofFedpolicy andgenerallyacceptedasthebasisofanticipa- decisionsonanumberofoccasions,mostrecently tions,anyoneofmanydifferentrules(orsetsof inaspeechonOctober4,2005,entitled,“How rules)wouldprobablyserveaboutaswellas PredictableIsFedPolicy?”Letmesummarize another.”4Ithinkhisfirstargumentwascorrect— themainfindings. thatdifferentrules,evenoncefullyunderstood, Overthepastdecade,theFederalOpenMarket wouldhavedifferentoperatingpropertiesinthe Committee(FOMC)hasundertakenanumberof economy,andthatachoiceamongvariouspossi- stepstowardgreatertransparencythathavegreatly blerulesshoulddependonwhichruleyields improvedtheabilityofmarketstopredictfuture bettereconomicoutcomes. policyactions.Amongthesestepsarethe Myviewhasevolvedovertimetothisgen- announcementofpolicyactionsattheconclusion eralposition:Monetaryeconomistshavenotyet ofeachFOMCmeeting;therestrictionofpolicy developedaformalrulethatislikelytohavebetter actionstoregularlyscheduledFOMCmeetings, operatingpropertiesthantheFed’scurrentprac- exceptunderextraordinaryconditions;the tice.Itishighlydesirablethatpolicypracticebe announcementofaspecificnumerictargetfor thefederalfundsrateinthepost-FOMCmeeting formalizedtothemaximumpossibleextent.Or, pressreleasesandintheDirectivetotheManager moreprecisely,monetaryeconomistsshould oftheopenmarketdeskattheFederalReserve embarkonaprogramofcontinuousimprovement BankofNewYork;theinclusionoftheindividual andenhancedprecisionoftheFed’smonetary votesattheFOMCmeetinginthepressrelease; rule.Itispossibletosayalotaboutthesystem- andtheexpeditedreleaseoftheminutesofthe aticcharacteristicsofcurrentFedpractice,even FOMCmeetings.Inaddition,since1989allFOMC thoughIdonotknowhowtowritedownthe policyactionstochangethetargetforthefunds currentpracticeinanequation.Itisinthissense ratehavebeeninmultiplesof25basispoints.With thatI’llbedescribingtheFed’spolicyrule.And theexceptionofonechangeof75basispoints, giventhat,asfarasIknow,thereisnoothereffort allthechangeshavebeeneither25or50basis tostateinoneplacethemaincharacteristicsof points. theFed’spolicyrule;I’msurethatsubsequent AsIhavenotedpreviously,Ibelievethatthe workwillrefineandcorrectthewayIcharacterize evidencesupportstheconclusionthatthesesteps therule.Thus,Iamredefiningthe“rule”tofit towardincreasedtransparencyhavebroughtthe currentpractice,whichhasyieldedanenviron- marketsintomuchbetter“synch”withFOMC mentinwhichpolicyactionsarehighly,though thinkingaboutappropriatepolicyactions.My notperfectly,predictableinthemarkets. metricforjudginghowwellmarketshaveantici- Beforeproceeding,Iwanttoemphasizethat patedFOMCpolicyactionsisthereactionofthe theviewsIexpressherearemineanddonot yieldonthe1-month-aheadfederalfundsfutures necessarilyreflectofficialpositionsoftheFederal contractbetweenthecloseofbusinessontheday ReserveSystem.Ithankmycolleaguesatthe beforetheFOMCmeetsandthecloseofbusiness FederalReserveBankofSt.Louisfortheircom- onthedayofthemeeting.Ourresearchsuggests 3 Simons(1936,p.5). 4 Simons(1936,p.29). 2 TheFed’sMonetaryPolicyRule thatchangesoflessthan5basispointsare cialshavestatedthatthegoaloflowandstable “noise.”Largerchangesreflectsurprisestomar- inflationistherebecauseitmaximizestheecon- ketexpectations. omy’ssustainablerateofeconomicgrowth.5 Sincethemiddleof1995,whentheFOMC ThedualmandateintheFederalReserveAct, hasundertakenpolicyactionsatregularlysched- asamended,andinotherlegislationprovidesfor uledmeetings,themarketshavebeensurprised goalsofmaximumpurchasingpower,usually only12times,asmeasuredbyachangeof5basis interpretedaspricestability,andmaximum pointsormoreinthe1-month-aheadfederalfunds employment.Therearetwoaspectstoachieving futurescontract.Sincethemiddleof2003,when theemploymentgoal.First,achievinglowand theFOMCintroduced“forwardlooking”language stableinflationmaximizestheeconomy’sgrowth intothepressrelease,therehavebeennosur- potentialand,probably,maximizesthesustainable prises.Incontrast,onallfouroccasionswhenthe levelofemployment.Second,theFedcanenhance FOMCinstitutedintermeetingpolicyactions,the employmentstabilitythroughtimelyadjustments marketsweretakenbysurprise. initspolicystance.Asubsidiarygoalofgeneral Ontheothersideofthecoin,FOMCdecisions financialstabilityiscloselyrelatedtobothinfla- toleavethefundsratetargetunchangedhavealso tionandemploymentgoals. becomelargelypredictable.Sincethemiddleof TheFedhasgravitatedtoaspecificationofthe 1995therehavebeenonlytwooccasionswhen inflationgoalstatedintermsofthecorepersonal themarketsexpectedachangeinthefundsrate consumptionexpenditures(PCE)index.Atthe targetandtheFOMCleftitunchanged. FOMCmeetingofDecember21,1999,Chairman Thesefindingsopenthisquestion:Whatare Greenspanprovidedaclearstatementofthecase thecircumstancesunderwhichmarketexpecta- forfocusingonthePCEpriceindexratherthan tionsofFOMCactionsareadjusted,sothat,bythe ontheconsumerpriceindex(CPI). timetheFOMCmeets,theoutcomesaregenerally correctlyforeseen?Thereisasubstantialliterature ThereasonthePCEdeflatorisabetterindicator inmyviewisthatitincorporatesafarmore documentinginterestrateresponsestoarriving accurateestimateoftheweightofhousingin information.Giventhatthefederalfundsfutures totalconsumerpricesthantheCPI.Thelatteris marketpredictsFOMCpolicydecisionsquite baseduponasurveyofconsumerexpenditures, accurately,thatliteratureprovidesinsightinto whichasweallknowverydramaticallyunder- howtheFOMCrespondstonewinformation. estimatestheconsumptionofalcoholand WhatI’lldonowistostepbackfromthatlevel tobacco,justtonameacoupleofitscompo- ofdetailtodiscusspolicyregularitiesatahigh nents.Italsodependsonpeople’srecollections level,startingwithpolicygoals. ofwhattheyspent,andwehavemuchharder evidenceofthatintheretailsalesdata,which iswherethePCEdeflatorcomesfrom.6 POLICY GOALS Thereisevidencethatthegoaliseffectively Onmanyoccasions,datingbacktoPaul a1to2percentannualrateofchange,averaged Volcker’sconfirmationhearingin1979,Fedoffi- overa“reasonable”periodwhoseprecisedefini- 5 See,forexampleCommitteeonBankingHousingandUrbanAffairs,UnitedStatesSenate,Ninety-sixthCongress,firstsession,Hearingsonthe NominationofPaulA.VolckertobeChairman,BoardofGovernorsoftheFederalReserveSystem,July30,1979,p.20;Committeeon Banking,HousingandUrbanAffairs,UnitedStatesSenate,Ninety-eighthCongress,firstsession,TheRenominationofPaulA.Volckertobe Chairman,BoardofGovernorsoftheFederalReserveSystemforatermof4yearsendingAugust6,1987,July14,1983,p.15;Committeeon Banking,HousingandUrbanAffairs,UnitedStatesSenate,OneHundredthCongress,firstsession,TheNominationofAlanGreenspanofNew York,tobeamemberoftheBoardofGovernorsoftheFederalReserveSystemfortheunexpiredtermof14yearsfromFebruary1,1978,vice PaulA.Volcker,resigned;and,tobeChairman,BoardofGovernorsoftheFederalReserveSystemforatermof4years,vicePaulA.Volcker, resigned,July21,1987,p.29;CommitteeonBanking,FinanceandUrbanAffairs,UnitedStatesHouseofRepresentatives,TestimonyofAlan Greenspan,February23,1988,reprintedintheFederalReserveBulletin,April1988,p.227. 6 FOMCTranscript,December21,1999,p.49. 3 MONETARYPOLICYANDINFLATION tiondependsoncontext.Evidencesupporting ing,thoughwecannotbeoverconfidentinour thisviewoftheinflationgoalappearsinthe abilitytoidentifysuchtendenciesandtodeter- minutesoftheFOMCmeetingsofMay6,2003, mineexactlytheappropriatepolicy andAugust9,2005.7 response.8 Iregardinflationstabilityastheprimarygoal Over10yearsago,JohnTaylor(1993)noted notbecauseitismoreimportantinawelfare thatthesecharacteristicsofFOMCpolicyactions sensethanmaximumemploymentbutbecause couldbesummarizedinasimpleexpression: achievinglowandstableinflationisprerequisite toachievingemploymentgoals.Inflationstability i = p + .5 ( p − p* )+ .5y + r* alsoenhances,butdoesnotguarantee,financial 1.5 p p* .5y r* p* , = ( − )+ +( + ) stability. Itakenote,butwillnotfurtherdiscusshere, whereiisthenominalfederalfundsrate,pisthe theongoingdebateastowhethertheinflationgoal inflationrate,p*isthetargetinflationrate,yis thepercentagedeviationofrealgrossdomestic shouldbeformalizedasaparticularnumerical goalorrange. product(GDP)fromatarget,andr*isanestimate ofthe“equilibrium”realfederalfundsrate. Underthischaracterizationofthesystematic or“rulelike”characterofFOMCpolicyactions, CHARACTERISTICS OF THE FED thefundsrateisraised(lowered)whenactual POLICY RULE inflationexceeds(fallsshortof)thelong-runinfla- TheFedpolicyrulehasanumberofelements tionobjectiveandisraised(lowered)whenoutput thatcanbeidentifiedand,inmanycases,quanti- exceeds(fallsshortof)atargetlevel.InTaylor’s fied.I’llnowdiscussthemostimportantofthese. example,thetargetforGDPwasconstructedfrom a2.2percentperannumtrendofrealGDPstarting The Taylor Rule withthefirstquarterof1984.Insubsequentanaly- StatementsandtestimonyofChairmen sesthistargethasbeeninterpretedasameasure VolckerandGreenspanandotherFOMCpartici- of“potentialGDP.”WheninflationandrealGDP pants,supplementedbythetranscriptsandmin- areon-target,thenthepolicysettingofthereal utesofFOMCdiscussionsoverthepast25years, fundsrateistheestimatedequilibriumvalueof clearlyindicatethatthelong-runobjectiveof therealrate.Thisformulationofaninterestrate FederalReservemonetarypolicyistomaintain monetarypolicyrulesatisfiesMcCallum’sprop- pricestability,usuallyphrasedas“lowandsta- ertiesforarulethatprovidesa“nominalanchor” bleinflation.”Intheshortrun,policyactionsare totheeconomy.9Taylorshowedthathisequation undertakenwiththeintentionofalleviatingor closelytrackedtheactualfederalfundsratefrom moderatingcyclicalfluctuations,asChairman 1987through1992exceptaroundthestockmarket Greenspanhasnoted: crashinOctober1987. Forsucharuletobeoperational,dataonthe [M]onetarypolicydoeshavearoletoplayover inflationrateandGDPmustbeknowntothe timeinguidingaggregatedemandintolinewith FOMC.Inpractice,theequationcanbespecified theeconomy’spotentialtoproduce.Thismay involveprovidingacounterweighttomajor, withlaggeddataoninflationandGDP.More sustainedcyclicaltendenciesinprivatespend- generallytheequationcanbewrittenasfollows: 7 FOMCMinutes,May6,2003,www.federalreserve.gov/fomc/minutes/20030506.htm;FOMCMinutes,August9,2005, www.federalreserve.gov/fomc/minutes/20050809.htm. 8 CommitteeonBanking,FinanceandUrbanAffairs,UnitedStatesHouseofRepresentatives,testimonyofAlanGreenspan,July13,1988. ReprintedinFederalReserveBulletin,September1988,p.611. 9 McCallum(1981). 4 TheFed’sMonetaryPolicyRule i p p* 100 ln y /yP r* p* , Policy Asymmetry t =a( t 1− )+ b ( t 1 t 1)+( + ) − − − Undermostcircumstancesthedirectionof – wherep isthepreviousquarter’sPCEinflation t–1 FOMCpolicyactionsis“biased”inasenseI’ll ratemeasuredonayear-over-yearbasis,y is explain.Policybiasexistsbecauseturningpoints t–1 thelogofthepreviousquarter’slevelofrealGDP, ineconomicactivity—peaksandtroughsofbusi- andyP isthelogofpotentialrealGDPasesti- nesscycles—areinfrequent.Changesineconomic t–1 matedbytheCongressionalBudgetOffice.To activityasmeasuredbyoutputandemployment ensurea“nominalanchor”fortheeconomy,the arehighlypersistent.Thispersistencecanbeseen coefficientamustbegreaterthan1.0. inFigure3,whichshowsmonth-to-monthchanges Figure1showstheequationwiththeTaylor innonfarmpayrollemploymentfromJanuary 1947throughAugust2005.Duringexpansions, coefficients(a=1.5,b=0.5),anassumedequilib- employmentchangesareconsistentlypositive; riumrealrateofinterestof2.0,andanassumed duringrecessionsconsistentlynegative.Changes inflationtargetof1.5percent.Thesolidblueline oppositetothecyclicaldirectionarerareand showstheactualfederalfundsrateandthedashed generallytheconsequenceofidentifiabletransi- linesthetwoTaylorrulefundsrates.Thesmall- toryshockssuchasthosefromstrikesandweather dashblacklineistheruleconstructedwiththe disturbances.Thispatternofbusinesscyclesgen- corePCEinflationrate;thelong-dashlightblue eratesstrongautocorrelationsinthemonth-to- linewiththePCEinflationrate.10Theaverage monthchangesinpayrollemployment,asshown differencesbetweenthetwo“Taylorrules”and inFigure4.11 theactualfundsrateovertheentireperiodare Givensuchpersistence,onceitbecomes 15and7basispoints,respectively.However,the apparentthatacyclicpeaklikelyhasoccurred, volatilityofeachofthetwoTaylorrulesismuch theissueisneverwhethertheFedwillraisethe lessthanthatoftheactualfundsrate. targetfundsratebutwhetherandhowmuchthe Figure2showsthecomparisonofthetwo Fedwillcutthetargetrate.Similarly,onceitis Taylorruleswithalargercoefficientontheoutput apparentthatanexpansionisunderway,theques- gap(b=0.8)andaslightlyhigherassumedequilib- tionisnotwhethertheFedwillcutthetargetrate, riumrealrate(r*=2.3).Withtheseassumptions buttheextentandtimingofincreases. theaveragedifferencesbetweenthetwoequations andthefundsrateovertheentireperiodare2and Data Anomalies –3basispoints,respectively,andthevolatilityof Fedpolicyrespondstoincominginforma- thetwoequationsbetterapproximatesthevolatil- tion,asitshould.Sometimesdataoughttobe ityoftheactualfundsrate. discountedbecauseofanomalousbehavior.For Mypurposehereisnottotrytofindtheequa- example,theFOMChasindicatedthatitmonitors tionthatrevealsthepolicyruleoftheGreenspan inflationdevelopmentsasmeasuredbythecore Fed;asIstatedearlier,Idonotknowhowtowrite ratherthanthetotalPCEinflationrate.This downthecurrentpracticeinanequationandthe approachisappropriatebecausetheimpactson FOMCcertainlydoesnotviewitselfasimplement- inflationoffoodandenergypricesarelargely inganequation.Rather,theillustrationsshould transitory;thedifferencebetweentheinflation beviewedasevidenceinsupportoftheproposi- rateasmeasuredbythetotalPCEindexandas tionthatthegeneralcontoursofFOMCpolicy measuredbythecorePCEindexfluctuates actionsarebroadlypredictable. aroundzero. 10TaylororiginallyspecifiedhisequationintermsofCPIinflation.SincetheFOMChasstatedapreferenceforPCEmeasuresofinflation, thosemeasuresareusedhere. 11AnestimateARIMAmodelformonthlychangesinnonfarmpayrollemploymentovertheperiodsince1947indicatesthat Payroll_Emp 0.96 Payroll_Emp 0.64 . ∆ t− ∆ t 1=εt− εt 1 − −− 5 MONETARYPOLICYANDINFLATION Figure 1 GreenspanYears: Funds Rate and Taylor Rules (p*= 1.5, r*= 2.0) a = 1.5, b = 0.5 Percent 12 Federal FundsRate TaylorRule–CorePCE 10 TaylorRule–PCE 8 6 4 2 0 Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 Figure 2 GreenspanYears: Funds Rate and Taylor Rules (p*= 1.5, r*= 2.3) a = 1.5, b = 0.8 Percent 12 Federal FundsRate TaylorRule–CorePCE 10 TaylorRule–PCE 8 6 4 2 0 Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 6 TheFed’sMonetaryPolicyRule Figure 3 Monthly Changes in Nonfarm Payroll Employment: January 1947–August 2005 1,250 1,000 750 500 250 0 –250 –500 –750 –1,000 1947 1952 1957 1962 1967 1972 1977 1982 1987 1992 1997 2002 NOTE:Shadedbarsindicaterecessions. Anotherexamplewastheincreaseintobacco areordinarilyrathereasytoidentify.BothFedand pricesinlate1998.Tobaccopriceshadatransitory marketeconomistsdevelopestimatesofthese impactonmeasuredinflation,forbothtotaland aberrationsinthedatashortlyaftertheyoccur. coreindices,duringDecember1998andJanuary Theprincipleoflookingthroughaberrationsis 1999,butproducednolastingeffectontrend easytostatebutprobablyimpossibletoformalize inflation.12Similarly,informationaboutreal withanyprecision.Weknowtheseshockswhen activitysometimesarrivesthatindicatestransitory weseethem,butcouldneverconstructacom- shockstoaggregateoutputandemployment.An pletelycomprehensivelistofsuchshocksexante. exampleofsuchatransitoryshockisthestrike Policymakerspiecetogetherapictureofthe againstGeneralMotorsinJuneandJuly1998.13 economyfromavarietyofdata,includinganec- Similarly,theSeptember2005employment dotalobservations.Whenthevariousobservations reportreflectstheimpactofHurricaneKatrina. fittogethertoprovideacoherentpicture,theFed Transitoryandanomalousshockstothedata canadjusttheintendedratewithsomeconfidence. 12FromtheDecember1998CPIreleaseinJanuary1999:“Three-fourthsoftheDecemberriseintheindexforallitemslessfoodandenergywas accountedforbya18.8percentriseintheindexforcigarettes,reflectingthepass-throughtoretailofthe45-cents-a-packwholesaleprice increaseannouncedbymajortobaccocompaniesinlateNovember.” 13FromtheJuly16,1998,FederalReserveStatisticalReleaseG.17IndustrialProductionandCapacityUtilizationpressrelease:“Industrial productiondeclined0.6percentinJuneafterarevisedgainof0.3percentinMay.Ongoingstrikes,whichhavecurtailedtheoutputofmotor vehiclesandparts,accountedforthedecreaseinindustrialproduction.”FromtheEmploymentSituation:July1998,releasedAugust7,1998: “Nonfarmpayrollemploymentedgedupby66,000to125.8million,asgrowthwascurtailedbystrikesandplantshutdownsinautomobile- relatedmanufacturing.” 7 MONETARYPOLICYANDINFLATION Figure 4 Autocorrelations of Monthly Payroll Employment Changes: January 1947–August 2005 1.00 0.75 0.50 0.25 0.00 1 2 3 4 5 6 7 8 9 10 11 12 Lag Themarketgenerallyunderstandsthisprocess,as withdrawnbeforeriskinganinflationproblem. itdrawssimilarconclusionsfromthesamedata. Intheabsenceofsuchconfidence,theFed’sability torespondwouldbeseverelycurtailed. Crisis Management ThehistoryofFedcrisismanagementsince WorldWarIIisgenerallyahappyone.Beforethe Theaboverulesaresuspendedwhenneces- Greenspanera,significanteventsincludethe sarytorespondtoafinancialcrisis.Themajor failureofPennCentralin1970andthenearfailure examplesoftheGreenspaneraarethestockmarket crashof1987,thecombinationoffinancialmarket ofContinentalIllinoisin1984.Perhapsjustas eventsinlatesummerandearlyfall1998that important,theFedhasnotrespondedtocertain culminatedinthenearfailureofLongTerm eventswhereitwascalledtodoso.Examples CapitalManagement,crisisavoidancecoming wouldincludetheNewYorkCityfinancialcrisis uptothecenturydatechangeattheendof1999, in1975andfailureofDrexelBurnhamLambert andthe9/11terroristattacks.Ineachcase,the in1990.14 natureoftheresponsewastailoredtocircum- Other Regularities in Policy Stance stancesuniquetoeachevent.Inallcases,crisis responseswerehelpfulbecausemarketshadcon- SinceAugust1989,theFOMChasadjusted fidenceintheFederalReserve,includingconfi- theintendedfederalrateinmultiplesof25basis dencethatextraprovisionofliquiditywouldbe pointsonly.AfterFebruary1994,whentheFOMC 14DrexelBurnhamLambertwasfirstinvestigatedbytheSecuritiesandExchangeCommissioninlate1987andchargedwithsecuritiesfraudin June1988.AsettlementwasreachedinDecember1988,butthefirmdeclaredbankruptcyinFebruary1990. 8 TheFed’sMonetaryPolicyRule firstbegantoannounceitspolicydecisionatthe scantyandIdonotbelievethatpolicystatements conclusionofitsmeeting,withfewexceptions arecurrentlyhighlypredictable. alladjustmentshavebeenmadeatregularly TwosignificantelementsinFOMCpolicy scheduledmeetings.Theseexceptionswere statementsarethe“balanceofrisks”assessment April18,1994,September29,1998,January3, introducedinJanuary2000andthe“forward 2001,April18,2001,andSeptember17,2001. looking”languageintroducedinAugust2003.The Ingeneral,theFedcanuseintermeeting balance-of-risksassessmentwasintroducedto adjustmentstorespondtospecialcircumstances, replacethelong-standing“bias”statementinthe suchastheratecutonSeptember17,2001,in DirectivetotheOpenMarketDesk.Historically, responseto9/11,ortoprovideinformationtothe thebiasstatementhadreferredtotheintermeeting marketaboutamajorchangeinpolicythinking periodandwasnotevenmadepublicintimely ordirection,suchastheratecutonApril18,2001. fashionuntilMay1999.Withtheregularization Myownpreferenceistoconfineintermeeting ofFOMCpolicyactionsonscheduledmeeting adjustmentstocircumstancesinwhichdelaying dates,andissuanceofastatementfollowing actiontothenextmeetingwouldhavesignificant everymeetingstartingwithMay1999toindicate costs.Ingeneral,ifthemarketbelievesthat whetherornotthefundsratetargetwaschanged, changedcircumstanceswillleadtoachanged aconsensusemergedamongFOMCparticipants thatthebiasformulationdidnotprovideaclear decisionatthenextregularlyscheduledmeeting, publiccommunication.Thebalance-of-risksstate- thenlittleisgainedbyactingbetweenmeetings. mentattemptedtoprovideinsightintothemajor Byreservingalmostallactionstoregularlysched- policyconcernsofFOMCmembersoverthe uledmeetings,intermeetingactionshavespecial “foreseeablefuture.” force,whichcanbevaluableinmeetingfinancial Initially,theCommitteesoughttosummarize crises. therisksforpolicyintheforeseeablefutureina singleassessmentcoveringtheprospectsforboth realeconomicactivityandinflation.InJune2003, ISSUES TO BE RESOLVED theassessmentoftheriskforsustainablegrowth Therules-versus-discretiondebatehistorically wasunbundledfromtheriskforinflation,allowing wasframedintermsofpolicyactions.Thefocus theCommitteetoexpressconcernsindifferent onpolicyactionswasnaturalbecause,historically, directionsaboutthetworisks.UntilApril2005, centralbankerswerereticenttocommentonthe thebalance-of-riskswasanunconditionalstate- rationalefortheirpolicyactionsandonlyrarely ment;sincethen,theassessmenthasbeencondi- providedhintsaboutthefuturecourseofpolicy tionedupon“appropriatemonetarypolicyaction.” actions.Overthepast15years,ascentralbankers, Overthe49FOMCmeetingssinceFebruary includingtheFOMC,havestrivenforgreatertrans- 2000,therehavebeen10substantivechangesin parencyinmonetarypolicy,communicationin thewordingofthebalance-of-risksstatement.15 theformofpolicystatementshasmovedtocenter Oneofthesechangeswasadecisionnottomake stage.Itisclearthatpolicystatementsarejustas abalance-of-risksassessmentonMarch18,2003, importantaspolicyactions,atleastintheshort inlightoftheuncertaintyassociatedwiththe run,becausesignificantmarketeffectscanflow Iraqwar.Intheremaining10formulationsofthe fromthesestatements.Weneedtofaceanew statement,5assessedtherisksasroughlybalanced question:Canpolicystatementsbecomepredict- (orbalancedconditionalonappropriatepolicy), able?Ithinktheanswerinprincipleislargelyin 3indicatedconcernabouteconomicweakness, theaffirmative,althoughevidenceontheissueis 1indicatedconcernaboutheightenedinflation 15ThesechangesoccurredonDecember19,2000;March19,2002;August13,2002;November6,2002;March18,2003;May6,2003;June25, 2003;December9,2003;May4,2004;andMarch22,2005. 9 MONETARYPOLICYANDINFLATION pressures,and1indicatedaconcernaboutthe haveprovidedconsistentsignalsaboutthedirec- riskthatinflationmightbecome“undesirablylow.” tionoffuturepolicyactions. TheswitchinlanguageonDecember19,2000, IninterpretingtheFOMC’spolicystatements, fromaconcernaboutheightenedinflationpres- itisimportantthateachstatementbereadagainst surestoeconomicweakness,wasfollowedbya previousones.Changesinthewordingarecritical reductioninthefederalfundstargetby50basis tounderstandingtheperspectiveoftheFOMC pointsatanunscheduledFOMCmeetingon membersaboutfuturepolicyactions. January3,2001.OnAugust13,2002,therisk assessmentwaschangedfrombalancedto weightedtowardeconomicweakness,butthe RULE ENFORCEMENT FOMCtooknopolicyactionsuntilitreducedthe Obviously,thereexistsnolegalenforcement targetforthefundsrateby50basispointsatits mechanismofthecurrentrule.Nevertheless, scheduledmeetingonNovember6,2002—thesec- therearecertainlyincentivesfortheFedChairman ondFOMCmeetingafterthechangeinlanguage. tofollowtherule,orworktodefineimprovements. Theriskassessmentwaschangedfrombalanced Themostpowerfulincentivesarisefrom toweightedtowardweaknessattheMay6,2003, marketreactionstoFedpolicyactions.Thefederal scheduledFOMCmeeting,andthefederalfunds fundsfuturesmarketprovidesasensitivemeasure ratetargetwasreducedby25basispointsatthe ofnear-termmarketexpectationsandtheeuro- subsequentFOMCmeetingonJune25,2003.Prior dollarfuturesmarketasensitivemeasureof toAugust2003,nopolicyactionswereunder- longer-termfundsrateexpectations.Thespread takenatagivenFOMCmeetingoritssubsequent betweenconventionalandindexedTreasurysecu- meetingwhentheriskassessmentwasbalanced. ritiesprovidesinformationoninflationexpecta- BeginninginAugust2003,theFOMCadded tionsor,moreaccurately,inflationcompensation. “forwardlooking”languagetothepressstatement. Optionsinthesemarketsprovideinformationon Initially,thelanguageindicatedthat“policy thediffusionofinvestorexpectations.Volatility accommodationcanbemaintainedforaconsid- ofmarketratesandaccompanyingmarketcom- erableperiod.”InJanuary2004,theCommittee changedthelanguagetoindicatethatitcouldbe mentaryprovidequickfeedbackastomarket “patientinremovingitspolicyaccommodation.” reactionstoFedpolicyactionsandpolicystate- TheFOMCdidnotchangethetargetfederalfunds ments.ItisnotintheFed’sinteresttoconfuseor ratewhilethesestatementswereineffect.InMay whipsawmarkets,andforthisreasonmarketreac- 2004,theCommitteeindicatedthatit“believes tionsprovideanincentivefortheFedtoconduct thatpolicyaccommodationcanberemovedata policyinapredictablefashionthatatthesame pacethatislikelytobemeasured.”Atitsfollowing timeachievespolicygoals.Policyactionsshould meeting,theFOMCraisedthefederalfundsrate beunpredictableonlyinresponsetoeventsthat targetby25basispoints.TheCommitteethen arethemselvesunpredictable.Theresponsefunc- raisedthetargetrateby25basispointsatallits tionitselfshouldbeaspredictableaspossible. subsequentmeetingsuptothetimethisspeech Thatis,giventhearrivalofnewinformation,the waswritten.Themostrecentsuchmeetingwas goalisthatthemarketshouldbeabletopredict September20,2005. thepolicyactioninresponsetothatinformation. Ataminimum,theFOMCcanandshould Althoughmarketresponsesarethemost aspiretopolicystatementsthatareclearanddo importantdiscipliningforce,FOMCmembers notthemselvescreateuncertaintyandambiguity. otherthantheChairmanalsoprovideinput, Therecordsince2000suggeststhatthebalance- includinginputthroughdissentswhenamember of-risksstatementandmorerecentlytheforward- feelsstronglythatadifferentpolicydecision lookinglanguageincludedinthepressreleases wouldbebetter.ReserveBankdirectorsweigh 10 TheFed’sMonetaryPolicyRule inthroughdiscountratedecisions.Since1994, dowithitsconstructionthananyoneelse.Never- exceptinunusualcircumstances,theFOMChas theless,Ibelievethatmostelementsoftherule notchangedtheintendedfederalfundsrateunless havebecomepartofageneralFedculture,under- severalReserveBankshaveproposedcorrespon- stoodatleastroughlybyotherFOMCmembers dingdiscountratechanges.16 andbystaff.Whileitisappropriatetorefertothe Finally,thegeneralroleofpublicdiscussion, “Greenspanrule,”IbelievethatFOMCdebates includingthehighlyvisiblecongressionalhear- andstaffcontributionshavehadalottodowith ings,bearsontheprocess.Skillfulpublicofficials developmentoftherule.Forthisreason,Ibelieve donotwanttobeforcedintoadefensiveposture thatweshouldbehopefulthatconsistentand whenconfrontingquestionsinhearingsandin predictableFedpolicyislikelytocontinueinto Q&Asessionsfollowingspeeches.I’llleaveitto thefuture. politicalscientiststostudythematterindetail, butwillguessthatpublicopinionplaysamore importantrolethanformallegalprocessesin REFERENCES enforcingmanylegislatedandcommonlawrules. Simons,HenryC.“RulesVersusAuthoritiesin Ifso,thenpublicopinioncanplayanimportant MonetaryPolicy.”JournalofPoliticalEconomy, roleinenforcingextra-legalrulesaswell. February1936,44(1),pp.1-30. McCallum,BennettT.“PriceLevelIndeterminacy A SUMMING UP withanInterestRatePolicyRuleandRational Expectations.”JournalofMonetaryEconomics, FederalReservepolicyhasbecomehighly November1981,8(3),pp.319-29. predictableinrecentyears;inthefuturethispre- dictabilitywill,Iamsure,beseenasoneofthe Poole,William.“HowPredictableIsFedPolicy?” hallmarksoftheGreenspanera.Littlehasbeen UniversityofWashington,SeattleOct.4,2005; institutionalized,andforthisreasonthecurrent www.stlouisfed.org/news/speeches/2005/ FederalReservepolicyrulemustberegardedas 10_04_05.htm;FederalReserveBankofSt.Louis somewhatfragile.Still,futureChairmenwillwant Review,November/December2005,87(6),pp.659-68. toextendAlanGreenspan’ssuccessfuleraand thereforeitwillbeintheirinteresttocommitto Taylor,JohnB.“DiscretionversusPolicyRulesin pursuepolicyregularitiesthatworkwell. Practice.”Carnegie-RochesterConferenceSerieson Idonotclaimtohaveaccuratelyidentified PublicPolicy.Amsterdam:North-Holland,1993, allaspectsoftheFed’scurrentpolicyrule.Iam 39(0),pp.195-214. temptedtocallitthe“Greenspanpolicyrule,” forAlanGreenspanhassurelyhadfarmoreto 16Duringthisperiodtherewere7occasionswhenthetargetfundsratewaschangedwithoutanaccompanyingactionbytheBoardof Governorstochangethediscountrate.Oftheremaining36changesintheintendedfundsrate,33wereaccompaniedbychangesinthedis- countrateatfourormoreFederalReserveBanks.On24oftheseoccasions,thediscountratewaschangedatamajorityoftheFederalReserve Banks. 11

William Poole (2005, October 13). Speech. Speeches, Federal Reserve. https://whenthefedspeaks.com/doc/speech_20051014_poole

@misc{wtfs_speech_20051014_poole,
  author = {William Poole},
  title = {Speech},
  year = {2005},
  month = {Oct},
  howpublished = {Speeches, Federal Reserve},
  url = {https://whenthefedspeaks.com/doc/speech_20051014_poole},
  note = {Retrieved via When the Fed Speaks corpus}
}