Speech

Monetary Policy Rules? TheJeffreyandKathrynColeHonorsCollegeLecture MichiganStateUniversity EastLansing,Michigan March16,1999 PublishedintheFederalReserveBankofSt.LouisReview,March/April1999,81(2),pp.3-12 All aspects of our behavior, economic ableway.”1Theconceptofamonetarypolicyrule and otherwise, are governed and istheapplicationofthisprincipleintheimple- organized by various rules. The most mentationofmonetarypolicybyacentralbank. routine aspects of our lives become Why,then,thequestionmarkinmytitle?There unpredictable and even chaotic when not gov- isalargebodyofeconomicsliteratureonthe erned by well-defined and generally accepted rules-versus-discretiondebateovermonetary rules. For example, in the 20th century, every policy.Idonotintendtoanalyze,orpsycho- country has adopted a legal rule that governs analyze,thisdebatehere.Iespeciallydonot automobile traffic. In the United States, it is intendtoaddressthepoliticalsideofthisdebate— understood that automobiles are driven to the whetheritwouldbedesirableforanationallegis- right;intheUnitedKingdom,theruleistodrive laturetoenactamonetarypolicyruletobe to the left. If, as a U.S. driver, you travel to executedbythecountry’scentralbank.Rather, England and are unaware of the rule for driving mypurposeistoexaminewhatwemeanbya or choose to ignore it, it is quite likely that the monetarypolicyrulefollowedbyacentralbank, outcome will be tragic. andtoexaminewhatweknowabouttheconstruc- Inasimilarvein,theconventionintheUnited tion,ordesign,oftherule. Statesistowalktotheright.Generallythisworks Iwillfirstdiscusssomegeneralissuesinthe well,andpedestriantrafficflowssmoothlyeven designofrules.Next,I’lldiscusswhatmonetary alongcrowdedwalkways.Insomecultures,the policycanachieve—weneedtobeclearabout rulesforpedestriantrafficdonotappeartobeas whatapolicyruleissupposedtoaccomplish. clearlyunderstood.Thereitiscommontoobserve I’llalsoreviewhowtheFederalReserveconducts considerableziggingandzaggingindensetraffic monetarypolicytodaytoprovidethebackground aspedestriansseektoavoidcollisionsinthe necessarytounderstandthepracticaldesignissues absenceofasystematicdecisionprocessonhow foramonetaryrule.Myfinaltopicwillbethe toproceed.Theabsenceofawell-understood criticalrequirementsthatmustunderlieanysatis- conventionforpedestriantrafficincreasestransit factoryrule.Thatwillbringmetoourcurrent timesandcanresultinconsiderableirritation. understandingofapracticalrule. Resources,whichcouldbeusedproductively, Beforediggingintothistopicfurther,however, arewasted. IwanttoemphasizethattheviewsIexpresshere Whatisarule?Arulecanbedefinedas reflectmythinkinganddonotnecessarilyreflect “nothingmorethanasystematicdecisionprocess officialFederalReserveviews.BobRasche,until thatusesinformationinaconsistentandpredict- recentlyamemberofthedepartmentofeconomics 1 AllanH.Meltzer,“Commentary:TheRoleofJudgmentandDiscretionintheConductofMonetaryPolicy,”ChangingCapitalMarkets: ImplicationsforMonetaryPolicy,FederalReserveBankofKansasCity,1993,p.223. 1 MONETARYPOLICYANDINFLATION atMichiganStateUniversity,isreallyaco-author Monetarypolicy,needlesstosay,ismorelike ofthislecture;hedeservescreditforitsstrengths pokerthansolitaire.Thegoalofmonetarypolicy, andI’llacceptresponsibilityforitserrors. however,istomaketheeconomybetteroff,not togohomeattheendoftheeveningwithyour friends’money.Thefactis,though,thatmonetary SOME GENERAL policyaffectsinterestrates,peoplemakeorlose moneyfrominterest-ratefluctuations,andthere- CONSIDERATIONS IN fore,themarketsareconstantlytryingtoforecast DESIGNING RULES thenextmonetarypolicyadjustment.Thisinter- Howarerulesconstructed?Inmanycases, action—policymakerstryingtounderstandand rulesgovernourinteractionwiththeenvironment, interpretmarkets,andmarketstryingtopredict andtheoptimalrulecanbeestablishedasthe whatpolicymakerswilldo—makesthetaskof solutiontoawell-definedengineeringproblem. designinganoptimalmonetarypolicyruleavery, Givenadvancesintechnology,manyrules,for- verydifficultproblem. merlyexecutedmanually,arenowperformedby automaticdevices.Forexample,thermostats thatcontroltheheatinourhomesandautopilots WHAT CAN MONETARY POLICY thatcontroltheprogressofaircraftembodysuch ACHIEVE? rules.Asecondtypeofrulegovernsourinterac- tionswithothers.Whentheabsenceofwell- Atthebeginningofthe1960s,economists definedrulesofbehaviororthefailuretoconform generallybelievedthatcentralbankscouldand toaconventionseriouslyharmsthewelfareof shouldbesignificantplayersintheeffortto others,thegovernmentimplementsarulebylaw achievemultiplesocialobjectives:lowinflation, andimposespenaltiesfordisobedience.Youwill highgrowth,lowunemploymentandlownominal beapprehendedquicklyifyouinsistondriving interestrates.Inaddition,theFederalReserve ontherightsideoftheroadinEngland. wasexpectedtocontributetospecificeffortssuch Athirdtypeofruleinvolvestheformulation asencouragingbalancedpaymentswiththerest ofpolicydecisions.Here,asystematicdecision- oftheworldandastronghousingsector. makingprocessiscomplicatedbecauseindivid- Thenotionthatcentralbankscanprovidea ualsandmarketparticipantsobserveorinferthe low-cost,over-the-counter“aspirin”thatwill actionsofthepolicymakersandadjusttheir alleviatealmostanyillthatasocietycanfaceis behaviorinwaysthatworktotheirbenefit,given nolongercredible.Thereisnowaconsensus theirunderstandingofthepolicyregime.This amongeconomistsandcentralbankersthatthe isthetypeofproblemfacedbymonetarypolicy onlylong-runeffectamonetaryauthoritycanhave decision-makers. onaneconomyistodeterminethesustained,or Thispointissoimportantthatitdeserves trend,rateofinflation.Thatratewillresultfrom specialemphasis.Comparetwocardgames: therateatwhichthemonetaryauthorityinjects solitaireandpoker.Solitaireisagameagainst moneyintotheeconomy. nature—thecharacteristicsofadeckofcards. Theviewthatpricestabilitydependsupon Thosecharacteristicsdonotchangefromoneday monetaryconditionshasalonghistoryinmone- tothenext.Pokerisagameagainstintelligent taryeconomics.Indeed,thebasicpropositionthat agents.Theplayersmaychangefromonedayto theamountofmoneydeterminesthepricelevel thenext.Playerslearnabouteachother’splay- originatedlongbeforeeconomicswasrecognized ingstylesovertimeandchangehowtheyplay. asadiscipline.Simplyput,Iwouldliketonote Designinganoptimalrule,orstrategy,forpoker thatinthelate19thandearly20thcenturies, isamuchmoredifficultproblemthandesigning economistswerepreciseaboutthenatureofthe anoptimalruleforplayingsolitaire. connectionbetweenmoneyandthegeneralprice 2 MonetaryPolicyRules? level.IrvingFisher,amongothers,madeimportant actionsinitiallyaffectoutput,unemployment, contributionstomonetarytheorylongbeforethe andrealinterestrates,eventhoughthelong-run GreatDepression.Thisidea—thatthegeneral impactontheserealvariablesisnil.Research pricelevelanditsrateofincreasedependspri- effortstoquantifytheseinitialeffects,however, marilyonthelevelofthemoneystockandits havefailedtoprovideprecisemeasuresofthe rateofincrease—felloutoffavorwiththeriseof impact,andatleastoneschoolofthoughtmain- Keynesiananalysisinthe1930sand1940s.The tainsthatsuchshort-runeffectsarenegligible. ideawasrevivedinthe1950sbyMiltonFriedman, Tojudgemonetarypolicyandcentralbankers, whohaslivedtowintheintellectualbattlethat wemustconcentrateonwhattheycanreasonably sustainedinflationiseverywhereandalwayswill beexpectedtoachieve,givenourcurrentstateof beamonetaryphenomenon. knowledge.Thereisacompellingcase,Ibelieve, Aconsensusalsoexiststhaterraticmonetary thatthesuccessorfailureofmonetarypolicymust policyhassometimesproducedinstabilitiesin bejudgedfirstandforemostbywhetheracentral theeconomy.MostanalystsnowagreethatFederal bankisabletoachievealow-inflationenviron- Reserveactionscontributedsignificantlytothe mentonasustainedbasis.Thatenvironmentis, severityoftheGreatDepressionintheUnited inturn,conducivetomaximumgrowthandeffi- States.Monetarypoliciescanmaketheeconomy cientutilizationoftheresourcesavailabletoa eithermoreorlessstable.Itisgenerallyacknowl- society.Highgrowthandefficientutilizationof edgedthatatleastsome—Ithinkalot—ofthe resourcesdependongovernmentpoliciesbeyond creditforthestabilityoftheU.S.economydur- thecentralbank’scontrol.Thatfact,however,does ingthepast15yearsisduetoFederalReserve notchangethepropositionthatacentralbank’s policyunderChairmenPaulVolckerandAlan contributionshouldbejudgedprimarilybythe Greenspan.Finally,itisgenerallyacceptedthat averagerateofinflation,andsecondarilybythe centralbanksareresponsibleforactingasalender stability,orlackthereof,oftheoveralleconomy. oflastresortintheeventofageneralizedliquid- Bythisstandard,thehistoryofthesecondhalf itycrisistomaintainthesoundnessandfunction ofthe20thcentury,intheUnitedStatesandin ofthepaymentsmechanism.Mosteconomists othercountries,isnotkindtocentralbankers.For accepttheviewthatpromptFederalReserve ashortperiodduringthelate1950sandearly actionsinOctober1987,afterthestockmarket 1960s,theU.S.economy(forallpracticalpur- crashandagainlastSeptemberaftertheRussian poses)experiencedpricestability.Thereturnto default,wereappropriatepolicyinterventions. lowinflationfollowingtheKoreanWarwascon- Ofcourse,wecontinuetodebatetheappropri- sistentwithpriorU.S.experiencethatinflation ateextentofFedactionstoalleviatealiquidity wasawartimephenomenon.Duringpeacetime, crisis—indeed,evenwhattrulyqualifiesasa citizensgenerallywereunconcernedabout liquiditycrisis—butthatdebatedoesnotdetract inflation. fromacceptanceofthegeneralprinciplethata Periodically,theGalluppollhasasked,“What centralbankresponseisdesirablewhenthecrisis doyouthinkisthemostimportantproblemfacing issevere. thecountrytoday?”Oneoftheresponsechoices Theproblemofdesigningmonetarypolicyto was:“Inflationorthehighcostofliving.”In1956, achievesustainedlowinflationandmore,rather severalyearsaftertheendoftheKoreanconflict, thanless,stabilityisfarfromtrivial.Weknow only13percentofrespondentsindicatedthis muchlessaboutthistaskthanweshould.Atthis concern.By1964,thefractionofrespondents point,noconsensusexistsonthesizeorreliability selectingthisresponsehaddroppedtosixpercent. oftheshort-runimpactofmonetarypolicyonan Sixteenyearslater,in1980,theeconomy economy.Aconsiderableamountofprofessional wasatthepeakoftheinflationthathadstarted opinion,thegeneralpopularfeeling,andfinancial- duringtheVietnamWar.Inflationasmeasured marketcommentaryholdthatmonetarypolicy bytheconsumerpriceindex(CPI)hadreached 3 MONETARYPOLICYANDINFLATION double-digits.Short-terminterestratesexceeded theirmortgagepaymentswithever-biggerpay- 20percent.Peoplecametounderstandthatthis checksmaysoonbeasquaintasatelephone inflationcouldnotbeattributedtowar;the withadial.Andretireeslivingoffintereston VietnamWarhadendedbythemid1970s,but certificatesofdepositandgovernmentbonds willbeshockedwhentheirsecuritiesmature inflationpersistedandindeedrosefortherestof toseehowfarinterestrateshavefallenasinfla- thedecade.ThefractionofGalluppollrespon- tionebbs.Already,ratesonsix-monthcertifi- dentswhorankedinflationastheNo.1problem catesofdepositaredowntoanaverageof4%, facingtheeconomyrosealmostcontinuously— accordingtoBanxQuoteInc.,andsomebig from27percentin1972,to47percentin1976, banksarepayinglittlemorethan1%onregular to61percentin1980.Allsegmentsofsociety savingsaccounts. sharedthisconcern. Today,thepricetranquilityofthelate1950s Clearly,oureconomy—indeedmanyaspects andearly1960shasbeenreestablished,bothin ofourbroadersociety—isdeeplyaffectedby theUnitedStatesandinmostotherindustrial inflationandbytheabsenceofinflation.The countries.IntheUnitedStates,annualinflation publichasnodoubt,andIhavenodoubt,thatthe duringeachyearfrom1983through1989was absenceofinflationisbetter.Acriticalquestion closeto4percent.Withtheexceptionofabrief facingallofusatthepresenttimeiswhetherthe increasejustbeforeandduringthePersianGulf inflationexperienceofthepast16yearswillbe conflictin1991,theinflationratehasdeclined sustained.Or,willthisperiodultimatelybe steadilytothepointwheretheeconomylastyear viewedinhistoryasawonderfulstrokeofgood wasclosetopracticalpricestability.By1993,only luck—ananomalyinanageofotherwisenearly 1percentofGalluppollrespondentsranked permanentinflation? inflationasthemostimportantproblemfacing thecountry.AnarticleintheWallStreetJournal lastmonthreflectedonasurveyconductedby HOW DOES THE FED CONDUCT YaleUniversityeconomistRobertShiller:2 MONETARY POLICY TODAY? Itisnowwidelyacceptedthathighratesof TheFederalReservehaspracticedaconsis- inflationcandamageaneconomybydistorting tentapproachtotheimplementationofmonetary markets,underminingpublicfaithingovern- policyatleastsincethemid1980s.Monetary mentandforcingallsortsofwastedeffort. policydecisionsaretheresponsibilityofthe Mr.Shiller’s1996surveyfoundthat84%of FederalOpenMarketCommittee(FOMC).The thepublic—thoughonly46%ofeconomists— FOMCconsistsofthesevengovernorsofthe feltthatpreventinghighinflationwasasimpor- tantaspreventingdrugabuseordeterioration FederalReserveSystem,thepresidentofthe ofschools.Areturnofinflationwouldbe FederalReserveBankofNewYork,andfourof jarring:ThelasttimetheLaborDepartment thepresidentsoftheremaining11regionalFederal checked,onlyoneinfivemajorunioncontracts ReserveBanks,onarotatingbasis.Thiscommittee includedanautomaticcost-of-livingadjust- meetseighttimesayear,todiscussthecurrent ment,downfrom60%intheearly1980s.A stateoftheeconomyandtheprospectsfornear- spreadoftheJapanese-styledeflationaryspiral, termdevelopments.Thecommitteethenvotes inwhichfallingpricesexacerbatearecession, oninstructions—theDirectivetotheSystemOpen wouldalsobepainful. MarketAccountManager—thatspecifyatarget Butrightnow,theU.S.hasneither.Instead, valueforthefederalfundsinterestrate.Thefederal itisexperiencinganabsenceofinflation.With- outinflation,theold-fashionednotionthat fundsrateistherateatwhichdepositoryinstitu- youngcouplescanbuyahouseandgrowinto tionsborrowandlendtoeachothertheirreserve 2 DavidWessel,“WithInflationTamed,AmericaConfrontsanUnsettlingStability.”WallStreetJournal,February22,1999,p.A.1. 4 MonetaryPolicyRules? balancesonthebooksoftheFederalReserve Animportantdevelopment—ifnottheimpor- Banks.TheFedusuallyreferstothetargetfederal tantintellectualdevelopmentthroughoutthe fundsrateastheintendedrate. past25yearsinourunderstandingofhowthe Oncetheseinstructionshavebeenapproved, macroeconomyworks—istherecognitionthat itistheresponsibilityofthestaffoftheOpen expectationsplayacentralroleinaffectingeco- MarketDeskattheFederalReserveBankofNew nomicbehavior.Previously,totheextentthat York,inconsultationwiththeChairmanandmem- expectationswereconsideredatall,theywere bersoftheOpenMarketCommittee,tokeepthe treatedinarathermechanicalfashion.Contem- actualfundsrateclosetotheintendedrate.The poraryanalysesnowpostulatethatindividuals DeskproceedsbybuyingandsellingU.S.govern- donotsimplylooktopasteconomicoutcomes mentsecuritiesfortheFederalReserve’saccount, toprojectthefuturepathofimportantconditions orbyengagingintransactionsthatarethepracti- liketheinflationrate.Instead,individualsunder- calequivalentofbuyingandsellinggovernment standthatitisintheirself-interesttocontemplate securities.WhentheAccountManagerdesiresto seriouslywhatpaththeFederalReservelikely offsetmarketforcesthataredrivingthefundsrate willpursueformonetarypolicyandtoaligntheir abovetheintendedratesetbytheFOMC,the expectationsaboutfutureinflationwiththeir Deskpurchasessecuritiesintheopenmarketfor perceptionsofFedactions. theFed’saccount.WhentheAccountManager Sucharoleforexpectationsisnotjustanele- desirestooffsetmarketforcesthataredrivingthe mentofelegantandstylizedeconomictheories. fundsratebelowtheintendedrate,theDesksells Expectationsinfluencemarketactivitiesdayin securitiesintothemarketfromtheFed’saccount. anddayout.Tradersinthefederalfundsfutures Thedirectresultofsuchpurchasesandsalesis contractsontheChicagoBoardofTrade,forexam- thattheamountofcurrencyand/orbalancesof ple,poreovertestimonyandspeechesofthe depositoriesattheFederalReserveBanksis ChairmanandFederalReserveofficials,searching increasedordecreased. forhintsaboutwhethertheFOMCwillchange Thisapproachtoimplementingmonetary theintendedfederalfundsrateatitsnextmeeting, policyisnotnew.Exactlythesameprocedures orsomemeetingafterthat.Financialmarketscan wereemployedduringthelate1960sandthrough- gyratewidelyinresponsetoaremarkwhoseinter- outthe1970s,theperiodofrisinginflation.There- pretationiscontrarytotheprevailingimpression. fore,thereisnoguaranteethatthetacticsof Amonetarypolicyrulemusttakeintoaccount monetarypolicy,ascurrentlypracticedbythe thesemarketexpectationsandspeculations.The FOMC,willbesuccessfulinmaintainingalow- goalshouldbethatinterestratesandothermarket inflationenvironment;theexactsameprocedures priceswillrespondtoobjectiveinformationabout deliveredtheVietnam-erainflation. theeconomy—thesameinformationthatmonetary policyitselfdependson.Thefactthatmarketsso oftenrespondtocommentsandspeechesbyFed officialsindicatesthatthemarketstodayarenot KEY DESIGN CONSIDERATIONS evaluatingmonetarypolicyinthecontextofa FOR A MONETARY POLICY RULE well-articulatedandwell-understoodmonetary Wemustaddresstwocriticalissuesinthe rule.Theproblemisadeepanddifficultone.The processofdesigningaruleformonetarypolicy. Feddoesnotknowhowtospecifyitsmonetary First,therulemusttakeintoaccountthefactthat policydecisionssothatthemarketcanlookat theindividuals’regardingtheFed’sfutureactions thesamedatatheFedlooksatandarriveatthe areanimportantdeterminantofeconomicout- sameconclusion.Imakethisstatementnotby comes.Second,therulemustbeveryexplicit wayofanycriticismofmyFedcolleaguesorstaff, abouttheinformationtheFOMCusestodetermine butsimplyasanhoneststatementofhowthings howtochangetheintendedfederalfundsrate. aretoday.Weapplyourbestjudgmenttothetask 5 MONETARYPOLICYANDINFLATION anddonotrelyonaformalrule,becausewedo establishaconstantrateofgrowthforthestock nothaveaformalrulewetrust. ofmoneyandmaintainthatgrowthratenomatter Mypointhereisthatacriticalpartofdesign- whatemergedfromthestateoftheeconomy. ingaruleisdealingwiththeinteractionofthe Friedman’sopponentsarguedthat,shouldthe Fedandthemarkets.Giventhisinteraction,an Fedadoptsucharule,itwoulddefaultonits importantfirstfeatureofanyrulemustbethatit responsibilitiestostabilizecyclicalfluctuations isformulatedinasystematicfashionandcanbe oftheeconomy.Theyfeltthatsuchstabilization communicatedeasilytothepublic.Arulethe requiredthattheFedexercisediscretioninthe publicdoesnotunderstandwillnotworksatis- conductofmonetarypolicy.Friedmancountered factorilybecausepolicychangesresultingfrom thathistoricallytheFedwastheprincipalcause applicationoftherulewillconstantlytakethe ofcyclicalfluctuationsintheeconomyandthat marketsbysurprise.Thepublicwillnotand muchofthedesiredstabilizationoftheeconomy shouldnotacceptaprocedurethatcreatespolicy wouldbeachievedifmoneygrowthwereconstant. changesthatseemtotallyunpredictableand, Nothinginthemodernconceptofamonetary therefore,arbitraryandcapricious. policyrulerequiresthattheFedpursueapolicy invarianttothestateoftheeconomy.Therestric- Second,apolicyrulemusthavethecorrect tionsimposedonFeddecision-makingbythe long-runproperties.Arulethat,iffollowedreli- monetary-ruleprocess,asdefinedhere,only giously,wouldpermitinflationtoriseorfallto requirethatdecisionstochangeornotchange unacceptablelevelswouldobviouslybedeficient. theintendedfederalfundsraterepeatedlyincor- Thisaspectofdesigningaruleis,fortunately, poratethesameinformationandrespondtothat relativelystraightforward. informationinthesameway. Third,Ithinkitisdesirable,thoughIconfess CriticsoftheruleapproacharguethattheFed substantialuncertaintyonthispoint,thatarule mustconsiderallavailableinformationaboutthe relyheavilyonthemarketitself.Onthewhole, economy.Thereisnothingintheconceptofa marketsdoagoodjobinallocatingresources monetaryrule,however,thatprecludestheFOMC efficientlyandmakingjudgmentsaboutthings fromreachingitsdecisionsbasedonawidevariety thataredifficulttopredict.Ithinkarulewillwork ofinformation.Allthatisrequiredisthatthesame bestifitcanestablishasolidandpredictablebase informationbeconsideredandincorporatedinto formonetarypolicy,leavingmaximumroomfor thedecision-makingprocessinthesamefashion marketstosetinterestratesandotherprices.For eachtimetheintendedfederalfundsratelevelis example,ifweknewofadirectwaytosettherate reassessed.Neitherthedataconsulted,northe ofinflationtozerodirectly,thenmarketinterest weightplacedonparticularpiecesofinformation, ratescouldbefreetoriseandfallascredit shouldbealteredinrepeateddecisions.Further- demandsriseandfall. more,operatingunderamonetaryrulerequires thatthebasisfordecidingwhethertochangethe intendedfederalfundsratebeclearlycommuni- MONETARY POLICY RULES catedtothepublic.Everyoneshouldbeableto makeinformedpredictionsaboutthefuturecourse NowI’mdowntothebottomlineofthislec- ofpolicy,givenknowledgeofthesamefactsabout ture—whatmightanactualmonetarypolicyrule thestateoftheeconomy.Therulecallingfora looklike? constantgrowthrateofthemoneystockhasmany Theplacetobeginiswiththepolicyrule desirablefeatures: advocatedbyMiltonFriedman,amongothers, startingduringthe1950s.Thelong-standing • Itiseasyforthepublictounderstand. controversyovermonetaryrulesderivesinlarge •Therateofinflationcannottakeofftoward partfromthisparticularrule.Friedman’spro- plusinfinityorminusinfinityifmoney posedrulewasthattheFederalReserveshould growthisheldconstant. 6 MonetaryPolicyRules? • Interestratesarefreetofluctuateinresponse ThesecondtermintheTaylorrulecallsfor tochangingmarketconditions. anadjustmenttotheintendedfederalfundsrate whentheinflationratedeviatesfromtheFOMC’s TheFriedmanrule,however,hasnotgained targetinflationrate.Continuingwiththeillustra- generalacceptance.Onereasonwhyisthatthe tionthatthetargetinflationrateis1percent,if term“money”mustbedefinedinanacceptable theactualinflationrateis2percent,thentheinfla- wayiftheruleistoworkandbeeasilyunder- tiondeviationis1percentagepoint.TheTaylor stoodbythepublic.Many,andperhapsmost, rulemultipliesthatdeviationbyaspecifiedcoeffi- economiststodaybelievethatchangesinthe cientandaddstheproducttotheintendedfederal amountofmoneydemandedbythepublicare fundsrate.Forexample,ifthecoefficientis1.5, ofsufficientsizeanddurationthatkeepingthe thentheinflationdeviationof1percentagepoint moneystockonasteadypathwilllikelyleadto yieldsanintendedfederalfundsratethatishigher muchlargerfluctuationsintheinflationrateand by1.5percentagepoints. levelofeconomicactivitythanwe’dlike.The Taylorhasemphasizedtheimportanceof betterwayofstatingthispointistosaythata havingacoefficientontheinflationdeviation centralbankusingitsbestjudgmentcanbeatthe termthatishigherthan1.0.Ifthecoefficientis performanceoftheFriedmanrule,andthatthis below1.0,thenanincreaseininflationwillcall claimiswelldemonstratedbytheFed’sperform- forthanincreaseintheintendedfederalfunds ancesinceitbegantoattackinflationin1979. ratethatissmallerthantheincreaseininflation. ThereareothercriticismsoftheFriedmanrule. Thatmeansthattherealrateofinterestwould Mypurposehere,however,isnottoreviewthis fallwheninflationrises,whichisarecipefora wholedebatebuttodiscusstheissueofrulesmore never-endingincreaseininflation.Everyoneagrees generally. thatthecoefficientontheinflationdeviationneeds Others—especiallyAllanMeltzerand tobeabove1.0,buthowmuchaboveisunknown atthistime.Therealsoisanissueofhowtodefine BennettMcCallum—haveworkedonvariantsof theinflationrate—whatindextouseandwhat theFriedmanrule.Thesearequantity-basedrules timeperiod.Usingtheinflationrateoverthelast thatyieldachanginggrowthrateofthemoney monthwouldintroduceagreatdealofrandom stockorthemonetarybase.Theresearchisprom- noiseintothefederalfundsratesetbytherule; isinganddeservesmoreattentionthanitgets. usingtheinflationrateaveragedoverthelastfive Anotherapproachisaninterest-raterule,in yearswouldyieldarulethatrespondstooslowly contrasttothequantityrulesjustdiscussed.The tochangingconditions.Theoptimallengthof policyruleforinterestratesthathasbeendis- theaveragingperiodisnotknownatthistime. cussedmostoftenforseveralyearsnowwaspro- Usinganinflationforecastmightbebetter,but posedbyStanfordeconomistJohnTaylorin1993. whoseforecast? Hisruleisanattractiveonetoconsiderbecause ThethirdtermintheTaylorruleisthedevia- itissocloselylinkedtotraditionalFedpractice tionofrealgrossdomesticproduct(GDP)fromthe insettinganintendedfederalfundsrate. pathofpotentialGDP.Wecancallthisquantity Taylorproposedthatthefederalfundsrate theGDPdeviation.Severalideasliebehindthis bedeterminedbyarulewiththreebasictermsin term.OneisthatiftheFedweretofollowamoney it.First,thefundsrateshouldequalanestimate growthruleinaneconomyinwhichtheproblems oftheeconomy’srealrateofinterestatazerorate withthatruledidnotexist,theninterestrates ofinflationplustheFed’stargetrateofinflation. wouldriseandfallascreditdemandsriseand Forexample,withanestimateofanequilibrium fallwiththestrengthoftheeconomy.Inthissense, realrateofinterestof2percent,andalong-run theTaylorrulemimicsthebehaviorofinterest targetrateofinflationof1percent,thebaserate ratesunderaconstant-money-growthrule.More forthefederalfundsratewouldbe3percent. generally,itseemssensiblethatiftheeconomy 7 MONETARYPOLICYANDINFLATION isbooming—runningwellabovepotential—then ofhistory,nomatterhowsuccessfulpolicywas interestratesshouldbesomewhathighertocheck duringthatperiod.Everyempiricaleconomistis theexcessivepressureonavailablelaborand alltoofamiliarwiththephenomenonofagreat capitalresources.Iftheeconomyisslack—oper- modelfitduringasampleperiod,followedby atingbelowpotential—theninterestratesshould utterdisillusionmentwiththeperformanceof besomewhatlowertoencouragegreaterutiliza- themodeloutsidethesampleperiod. tionofavailableresources. Asyoucanimagine,therearenumerous Theargumentforaterminthepolicyrule potentialproblemswiththeTaylorrule;I’ve reflectingtheGDPdeviationisattractive.The mentionedjustafew.Buttheseproblemsalsoare optimalsizeofthecoefficientonthistermisnot problemswiththecurrentconductofmonetary known,however.Whetherthefederalfundsrate policy.Gaugingthecurrentinterestrateagainst shouldchangebyanamountequalto0.5,or2.0, theequilibriumrealrateofinterestisanimpor- orsomeothernumbertimestheGDPdeviation tantpartofourjobtoday,butwedon’tknowwhat isbeinginvestigated.Moreover,potentialGDPis theequilibriumrealrateisinprecisenumerical notanobservablevariable. terms.Weneedtojudgethecurrentlevelofthe Althoughdifferentresearchershavedifferent economyagainstitspotential,butwedon’tknow ideasaboutwhichisthebestmethodofmeasuring inprecisenumericaltermswhatthatpotential potentialGDP,Idonotregardthisissueascritical, is.CriticizingtheTaylorrule,oranyotherrule, becauseanerrorherewillnotsendtheeconomy forsuchreasonsdoesnothelpsolvetheproblems offpermanentlyinonedirectionortheother.The policy-makersface,nordoesitmakeaconvincing rulewillbestabilizing,thoughnotperfect,if casethatwhatwedonowisbetterthanfollowing potentialGDPismisestimated.Thesizeofthe animperfectrule. coefficientontheGDPdeviationisanimportant TheTaylorrulehasfiguredmoreprominently issue,however,asitdetermineshowstabilizing thanotherproposedrulesinrecentdiscussions theruleislikelytobe.Thisisacomplicated ofmonetarypolicyrules.Undoubtedly,interest matter—toolargeacoefficientmightinduce intheTaylorrulereflectsthefactthatthecurrent economiccyclesaroundpotentialGDPandtoo implementationofmonetarypolicyaroundthe smallacoefficientmightpermitsustaineddepar- globefocusesonmanipulatingashort-terminter- turesfromadesirablepaththatwouldtendto estrate.Thus,theadoptionofaTaylor-typerule destabilizetherateofinflation. wouldnotrequiretheFederalReservetoalterits Taylorarguedthatthebehaviorofthefederal operatingprocedures;theTaylorruleisaneffort fundsrateincorporatedinhisruleisareasonable toformalizethedecision-makingprocessthat approximationtotheactualprocessofadjustment generatestheoperatinginstructionsintheFed’s ofthefundsratetargetstheFOMCusedbetween currentpractice. 1987and1993—aperiodduringwhichmonetary Anotherapproachtomonetarypolicy,known policywasquitesuccessful.Buthisruleismuch asinflationtargeting,hasbeeninstitutedbythe morethanasimpleefforttofitthedata.Itincor- centralbanksinseveralforeigncountries.3The poratesimportantandsoundtheoreticalprinciples practice,whichvariesfromcountrytocountry, thatneedtobefollowedifmonetarypolicyisto startedwiththeReserveBankofNewZealand, besuccessful.AsIhavealreadyemphasized, andhasbeenadoptedbytheBankofCanada,the however,thereismuchwedonotunderstand BankofEngland,theBankofFinland,theSwedish abouttheoptimalconstructionofamonetary Riksbank,andtheReserveBankofAustralia. policyrule.Weshouldbewaryaboutaccepting Thesecentralbanksannounceinadvancetheir coefficientsthatseemtocomeoutofasmallslice policyobjectiveforaninflationrate.This 3 GuyDebelle,PaulMasson,MiguelSavastano,andSunilSharma,“InflationTargetingasaFrameworkforMonetaryPolicy,”Economic Issues15,InternationalMonetaryFund,Washington,D.C.,September1998. 8 MonetaryPolicyRules? announcementreflectsapubliccommitmentof itseemstomethatweignorethebehaviorofthe thecentralbanktothepolicyobjective.Innone monetaryaggregatesatourperil. ofthesecaseshasthecentralbankspecifiedthe Obviously,interestratesreflectmarketexpec- decisionrulethatitwillusetoachievethestated tationsaboutthefuture.Mightweincorporate objective.Innomarketeconomycanthecentral ratesintheruleinsomefashion?Theideais bankcontroltheinflationrateorthepricelevel intriguing,butthereisacircularityproblem becauseitappearsthatthebondandmoney directly.Itmustinterveneinsomemarketto marketsrespondsignificantlytochangesinFed manipulateapriceoraquantitytosteertheecon- policyandtochangesinexpectationsaboutFed omytowardthedesiredobjective—whetherthis policy.Themoreconfidencethemarkethasin objectiveisannouncedornot.Thevariablemanip- theFed,themorethemarketwillconcentrateon ulatedbythecentralbankisdefinedasthepolicy whattheFedisdoingandthelessthemarketwill instrument.Themonetaryruleindicatesthe concentrateonfundamentalsotherthantheFed. processbywhichthecentralbankadjuststhe Considerananalogy:Ifyouareaninvestorbut policyinstrumentwheninformationontheper- knowlittleaboutinvesting,itmakessenseto formanceoftheeconomyrelativetoitspolicy chooseyourinvestmentsbyobservingthedeci- objectiveisreceived.Thus,merelyannouncinga sionsofaninvestorknownforhissuperiorper- policyobjectiveintermsoftheinflationratedoes formance.Thisstrategyislikelytoworkbetter notassurethatthecentralbankisoperatingunder foryouthanconcentratingoninvestmentfunda- amonetarypolicyrule.However,Ibelievethatit mentalsthemselves,whichbyassumption,you isdesirableforcentralbankstobeclearabout maynotunderstandverywell.Themarket theirobjectives,andinthisregardIbelievethat watchestheFedbecausetheFediswellinformed inflationtargetingisadesirablepractice. andbecausetheFedisthedominantplayerin themoneymarket. Themoreconfidencethemarkethasinthe Fed’swillingnesstodowhateverisnecessaryto THE PATH TO PROGRESS IN maintainlowinflation,themoresenseitmakes MONETARY POLICY DECISIONS forthemarkettoconcentrateontheFed’sactions IbelievethattheTaylorruleisapromising ratherthanforminganindependentjudgment approachtobetterunderstandingmonetarypolicy. aboutfutureinflationprospects.Therefore,the FromwhatIhavealreadysaid,however,itshould Fedcannotusethebehaviorofinterestratesin beclearthatIdonotbelievethatthisruleisready thebondmarkettoprovideusefulinformation onhowitshouldadjustthefederalfundsrate. foradoption.Weneedalotmoreresearch. If,however,theFedisabletoadoptaprecise Twoareasdeservespecialattention.One ruleinthefuture,whichisbasedoninformation concernsaroleformonetaryaggregatesandthe everyonecanobserve—suchasthatemployedin secondaroleformarketinterestrates,orsome Taylor’srule—thentheremaywellbeaplacefor otherpieceofmarketinformation. aninterestratetermintherule.Ifthemarketcould Asageneralproposition,wecanextractinfor- beconfidentthattheFedwouldchangethefed- mationfrommarketsbystudyingbothpricesand eralfundsrate,onlyinresponsetotheobserved quantities.Thetotalneglectofinformationabout inflationrateandtheGDPdeviation(andwhat- themonetaryaggregatesintheTaylorruleopens everotherobservableinformationtheFed upanaturalavenuetoextendtherule.Thisis includedintherule),thenmarketforecasts nottheplacetopresentcurrentresearchresults, wouldbeincorporatedininterestrates.Adding especiallysinceIdon’thaveanyofthiskind.Iam aninterestratetermtotherulewouldbeaway simplysayingthat,givenalltheevidencesupport- toaddforecaststotherule,which,inprinciple, ingMiltonFriedman’spropositionthat“inflation shouldmakeitworkbetterthanarulebased isnowandeverywhereamonetaryphenomenon,” solelyonpastdata. 9 MONETARYPOLICYANDINFLATION CONCLUDING REMARKS shouldbeincorporatedintotheformulationof futurepolicyrules.Possiblechangesintherule Theconceptofamonetaryruleisattractive shouldbestudiedanddebated.Whentheanalysis formanyreasons.Torepeatthedefinitionoffered indicatesthattherulecanbeimproved,theFed atthebeginningofthislecture,aruleis“nothing shouldannouncethechangesinadvanceand morethanasystematicdecision-makingprocess explaintherationaleforthem. thatusesinformationinaconsistentandpre- Apolicyrulealsoprovidesthesurestmethod dictableway.”Operatingunderamonetaryrule topasstheaccumulatedknowledgeaboutthe imposesaccountabilityandtransparencyupona effectiveoperationofmonetarypolicytofurther centralbank.Itrequiresthatthepolicymakersbe generations.This,afterall,ishowengineers,using specificabouttherationalebehindtheirpolicy engineeringtheoryandobservationofskilled actions.Therecordofthedecisionswillthen humanpilots,constructedautopilots.These containinformationfromwhichfuturedecision- deviceshadtobetestedandrefined,andthe makerscanlearn. processtooktime.Thedevicesaresubjectto Apolicyruleoughtnotbeconsideredirrevo- continuousimprovement.Undernormalflying cableorunchangeable.Atanytime,ourunder- conditions,anautopilotnowdoesabetterjob standingoftheshort-runimpactofmonetary thanahumanpilotatkeepinganaircraftonits policyontheeconomyisimperfect.Policymakers desiredcourse.AsIhaveemphasized,designing necessarilyoperatewithinconstraintsimposed amonetaryruleisadifficulttaskintellectually, bythecurrentstateofknowledgeandshouldnot butitseemsobvioustomethatthisisthepath beblamedforoutcomesthatareimpossibleto wemusttravel. avoidgiventhatknowledge.Oneofthebenefits Inshort,pursuingthepathtodeveloping ofapolicyruleisthatahistoricalrecordwill andthenadheringtoaruleprovidesthebest existthatcanbeanalyzed,andfromthisanalysis approach—perhapstheonlyapproach—to wecanobtainanunderstandingofwhypastpolicy improvingthepracticeofmonetarypolicyover actionsdidnotproducetheintendedresults.The thelongrun. knowledgegainedfromsuchanalysiscanand 10

William Poole (1999, March 15). Speech. Speeches, Federal Reserve. https://whenthefedspeaks.com/doc/speech_19990316_poole

@misc{wtfs_speech_19990316_poole,
  author = {William Poole},
  title = {Speech},
  year = {1999},
  month = {Mar},
  howpublished = {Speeches, Federal Reserve},
  url = {https://whenthefedspeaks.com/doc/speech_19990316_poole},
  note = {Retrieved via When the Fed Speaks corpus}
}