speeches · December 4, 2002

Speech

William Poole · President

Institutions for Stable Prices: How To Design an Optimal Central Bank Law FirstConferenceoftheMonetaryStabilityFoundation RegionalOfficeoftheDeutscheBundesbank Frankfurt,Germany December5,2002 PublishedintheFederalReserveBankofSt.LouisReview,September/October2003,85(5),pp.1-5 Iam pleased to be here today to discuss an ForsomeyearsafterWorldWarII,mostobservers extremely important topic. But I believe it believedthatfixedexchangerateswereessential wise to begin on a humble note. The title tomonetarystability.Clearly,popularopinion ofthissessionincludesthephrase“optimal andunderstandingofeconomicideasimposes central bank law.” In designing a central bank limitsonourabilitytotransformtheeconomy law, we do not have a well-specified mathemat- bychanginglaws. ical model to optimize, and consequently we Beforeproceeding,Iwanttoemphasizethat cannot expect to find the optimal law. It would theviewsIexpressherearemineanddonot be a mistake, I believe, to be so bold as to rec- necessarilyreflectofficialpositionsoftheFederal ommend a legal framework for all countries for ReserveSystem.Ithankmycolleaguesatthe all time. The fact is that most high-income coun- FederalReserveBankofSt.Louisfortheircom- tries today, and many low- and middle-income ments,butIretainfullresponsibilityforerrors. countries, have achieved a high degree of suc- I’llorganizemythoughtsinfoursections.In cess in maintaining low inflation, even though thefirst,verybriefsection,I’lldiscusseconomic lawsinthesecountriesdisplaysubstantialdiffer- principles.Istarttherebecausethelegalframe- ences. We need to think rather abstractly about workwithinwhichacentralbankoperatesmust the design of the legal framework for the central beconsistentwiththewayamarketsystemworks, bank and recognize that there are different ways andthegoalsassignedtoacentralbankmustbe to achieve the same end. withinitspowertoachieve.Next,I’lldiscuss Weshouldalsorecognizethatsuccessin centralbanklawconsistentwitheconomicprin- achievinglowandstableinflation,orpricestabil- ciplesand,inaseparatesectionbecauseofits ityifyoupreferthatformulation,isrelatively importance,thedesignofcentralbankindepend- recent.Wemaywelldiscoverthatsomeinstitu- ence.Finally,I’lladdresstheissueofcentralbank tionalarrangementsaremorerobustovertime, transparency. asweobservehowvariousarrangementsstand Tomaketheexpositionabiteasier,I’llreferto uptostressesnotyetobserved. Aninstitutionasimportantasacentralbank theleadershipofacentralbankasthe“governor,” cannottakeaparticularformwithoutsubstantial whichwillrefertothegovernor,chairman,or publicunderstandingofthereasonsforthatform. governingboardasappropriate.I’llrefertothe Acenturyago,mostinformedpeoplebelieved topelectedofficialofthegovernmentasthe thattheonlysoundbasisforamonetarysystem “president,”whichwillrefertothepresidentor wasforpapermoneytobeconvertibleintogold. primeministerasappropriate. 1 MONETARYPOLICYANDINFLATION ECONOMIC PRINCIPLES Ipersonallyfavoralegislatedinflationtarget, butwhetherthetargetislegislatedisnotthemain Thelogicalplacetobeginananalysisofhow issue.Iftheweightofpublicopinionisnotbehind todesignanoptimalcentralbanklawiswitha alegislatedtarget,itwillnotbeeffective.The simplestatementofeconomicprinciples.The UnitedStatesdoesnothavealegislatedtarget, principlesIbelieveshouldguideourthinking butsincethemid-1990stheFederalReservehas arethese: beensuccessfulinachievingandmaintaininga lowaveragerateofinflation.Whatisneededis • Inflation,anticipatedandespeciallyunan- notsomuchalegislatedinflationtargetbutatarget ticipated,abovesomethresholdrateis frameworkthatthepublicregardsashavingcon- costly.Deflationisalsocostly.Costsare stitutionalforce.IntheUnitedStates,thegold lowwhenthedepartureoftherateofprice standardusedtohaveconstitutionalforceeven change,whetheraboveorbelowthethresh- thoughitwasneverwrittenintotheConstitution oldrate,issmall;costsarelargerwhenthe explicitly. departureislarger.Theevidencesuggests WhatImeanby“constitutionalforce”isthat thatthecostsofdeparturesarenotsymmet- alaworpracticecannotbechangedwithoutresort ric;deflationof5percentperyearislikely tolengthydiscussionand,inthecaseofalaw,by tobemuchmorecostlythaninflationof5 asupermajorityoritsequivalent.Aprovisionof percentperyear. constitutionalforceisbasictothefunctioningof • Thereisnolong-runtradeoffbetweeninfla- society;itispartofthesharedconsensus,backed tionandunemployment,andtheshort-run bywidespreadconsent,withinwhicheveryday legislationiscrafted. tradeoffmaywellbetoounreliabletobe IntheUnitedStates,repealofFirstAmend- usefulforpolicymakers. mentprotectionoffreedomofspeechisunthink- • Marketexpectationsaboutfuturemonetary able,andthatwasessentiallythesituation policy(andfutureeconomicpoliciesgen- applyingtothegoldstandardformanyyears. erally)areextremelyimportantindetermin- Butwhenthegoldstandardceasedtohavecon- inghowwellmonetarypolicywillwork. stitutionalforceasaconsequenceoftheGreat Depression,overtimeCongressrepealedlegisla- tionprovidingforgoldcoinsandotherfeatures CENTRAL BANK LAW ofthegoldstandard.Ithinkitistrue—Ihopeit istrue—thatintheUnitedStatestodaytheidea Becauseinflationanddeflationarecostly,a thatCongressortheFederalReservewoulddelib- centralbankoughttohaveaninflationtarget.I eratelyaimfor,ortolerate,asustainedinflation believethattheappropriatetargetiszeroinflation, rateof,say,8percentperyearisnowunthink- properlymeasured—thatis,abstractingfrom able.Ifso,theideathattheFederalReservehas measurementerrorsinpriceindexes.Others aresponsibilitytomaintainlowandstableinfla- believethatasmall,positiverateofinflationis tionintheneighborhoodofrecentexperienceis appropriate.Thedifferencebetween0and,say, approachingthelevelofconstitutionalforce. 2percentinflationperyearisaminormatter Iamsure,however,thatinmanycountries relativetootherissues.Inparticular,reasonable debateoveralegislatedinflationtargethasbeen stabilityintherateofinflationandespeciallyin extremelyvaluableinhelpingtocreateaconsen- theexpectedrateofinflationoverthemedium susofconstitutionalforce.WhatIamemphasiz- termaremoreimportantthanwhetherthetarget ingisthatsuchlegislationcanneverbetheend rateis0or2percentperyear.Whetherthetarget ofthematter;centralbankersandothersmust isexpressedasapointorarangeisaninteresting constantlyexplainthereasonsforalegislated issue,butnotfundamental. targettoensurethatitisnotsimplyabsorbed 2 InstitutionsforStablePrices:HowtoDesignanOptimalCentralBankLaw intotheimmensemassoflegislationonthebooks intheUnitedStatessince1982inthisway;most ofourdemocraticcountriesthatiswidelyignored recently,IthinkthatitisundeniablethattheFed’s andlargelyforgotten. rapidreductioninitsfederalfundsratetargetin Becausetheeffectivenessofacentralbankin 2001helpedtosoftentheextentoftherecession. achievingsustainedlowinflationdependsimpor- Ofcourse,wecannotjudgethesuccessofapol- tantlyonitscredibility,thereisnosubstitutefor icybyoneincompleteepisode—thejudgmentof consistentpoliciesthatbuildmarketconfidence historymightbethatpolicywastooeasytoolong, overtime.Oncecredibilityislost,regainingit althoughthatiscertainlynotmyjudgmentat takestimeandawillingnesstoendureshort-run thistime. pain,wheretheshortrunmaybemeasuredin Mypointisnottooffercommentaryon years.Maintainingcredibilityovertimerequires recentFederalReservepolicybuttoemphasize institutionalstrengththattranscendscurrent thatsuccessontheinflationfrontprovidesthe leadership.Absentcrisisconditions,policyshould opportunitytoemploymonetarypolicytostabi- evolverelativelyslowlyovertime,witheach lize,ortoworkinthedirectionofstabilizing, changestudiedcarefullyandthenexplained short-runfluctuationsinrealactivity.AndifI fully.Otherwise,thepredictabilityuponwhich amcorrectthatacentralbankthatissuccessful credibilitydependsmaybeincomplete.Thepur- ontheinflationfronthasthepowertocontribute poseofsustainedlowinflationistominimize toeconomicstability,thenIseenoreasonwhya pricelevelshocksthatupsetbusinessplanning governmentshouldnotassignacentralbankan andredistributeincomeandwealtharbitrarily. objectiveofcontributingtostabilityofthereal Forthesamereason,thecentralbankshouldstrive economytotheextentconsistentwiththeinflation toavoidsurprisesinitsownpolicyprocedures. objective.TheFederalReserveoperatesundera Oneofthemostdifficultandhotlydebated vaguelegislatedinstruction—vagueinthesense issuesiswhethermonetarypolicyshouldbe thatnonumericaltargetsarespecified—tocon- confinedtoaninflationobjectiveorshouldalso tributetoachievinghighemploymentandprice haveanemploymentorgrowthobjective.Myview stability.Ifthestatutorylanguageisinterpreted isthatitdoesnotmakeeconomicsenseforthe asIhavesuggested,thenIthinksuchobjectives centralbanktohaveobjectivesstatedintermsof makeperfectlygoodsense. thelevelofemploymentortherateofgrowthof Alegislatedemploymentstabilizationobjec- realgrossdomesticproduct(GDP).Itiswithinthe tivecomplicatestherelationshipbetweenthe powerofthecentralbanktoachievealong-run electedgovernmentandthecentralbankbecause inflationobjective,butnottoachieveanobjective thecentralbankmustmaintainalonghorizon. forthelevelofemploymentortheunemployment rate.Noorganizationshouldbeassignedanobjec- Thathorizonistypicallyconsiderablylongerthan tivethatitcannotachieveor,atbest,achieveonly thehorizonofelectedofficialswhoquitenatu- temporarily. rallyandunderstandablyhaveanintensefocus Ithinkitiswithinthepowerofthecentral onthenextelection.Becauseofthewaytheecon- bank,however,tocontributetoemploymentsta- omyworks,acentralbankmustbewillingtoback bility.Ifinflationexpectationsaresolidlyheld, awayfromeffortstostabilizeincomeandemploy- whichisanexpectedoutcomeofachievingan mentwhensucheffortsthreatentheinflation inflationobjectiveonasustainedbasis,thenthe objective.Failingtomaintaintheprimacyofthe centralbankcanreliablychangerealinterestrates inflationobjectiveonlyputseconomicstability intheshortrun.Providedthatthecentralbank’s atriskoverthelongerrun.TheUnitedStates short-runpolicydecisionsdonotshakeconfi- andmanyothercountrieshadampleexperience denceinthelong-runpolicy,itcandirectshort-run withthisscenariointhe1970s;excessesinshort- policytohelpcushionemploymentfluctuations. runrecessionfightingcreatedhigherinflation Itisreasonabletointerpretanumberofepisodes overthelongerrunanddeeperrecessionslateron. 3 MONETARYPOLICYANDINFLATION Centralbankindependenceistheinstitutional difficultoneforademocraticsociety:Howcan designthatpromisestoreconcilethedifferent animportantareaofpublicpolicybeofflimits horizonsofelectedofficialsandthecentralbank. forcommentandcriticismbyelectedofficials? Thissubjectissoimportantthatitdeservesspe- Yet,suchcriticismclearlyunsettlesmarketsand cialattention. damagestheeffectivenessofmonetarypolicy. Theonlywayaroundthisproblem,itseems tome,isforthegovernmenttoexercisegreatfor- CENTRAL BANK INDEPENDENCE bearanceandconfinecriticismtointernaldis- cussionswiththecentralbank.Thathascometo Thereiswidespreadagreementthatcentral bethepracticeintheUnitedStates,butithas bankindependenceleadstobettermonetarypol- notbeenestablishedlongenoughthatitcanbe icy.I’veintroducedthelogicofindependenceby regardedasinstitutionalized.Considerationof referringtodifferenthorizonsofelectedofficials thisissuemakesclearthatoptimalcentralbank andcentralbanks,butI’mnotsurethatisthetotal designgoesfarbeyondlegalissuesperse;itis story.Electedofficialsdomaintainsomepolicies ludicroustoconsiderthepossibilityofpassinga withgreatcontinuityovertimeandmakesome lawsayingthatthepresidentisnotallowedto investmentswithlongpaybackperiods.Fortwo commentoncentralbankpolicy!Clearly,though, quitedifferentU.S.examples,considerthelong ifthepresidentdoesnotretainconfidenceinthe horizonbehinddecisionstoinvestinnational centralbank,thecountryisinsubstantialtrouble. parksandmilitaryresearch. Inthissituation,thepresidentmustbeprepared Inote,however,thatcompetitionamongthose toreplaceafailingcentralbankleadershipwhen seekingelectoralofficedoesnotworkwellinthe termsexpire. contextofcentralbankleadership.Democratic Centralbankingisagovernmentalfunction, leaderscompeteforofficepromisingchangeand butIthinkthatsomeobserversmostcommitted improvementratherthancontinuityandstability, todemocraticprinciplesoverlookthepossibility whereasanincomingcentralbankgovernorwill ofemployingprivate-sectoractivityandprinci- almostcertainlywanttocontinuethepoliciesof plesforgovernmentalends.Awell-understood asuccessfulpredecessorandwillemphasizehis exampleisthevalueofusingpollutiontaxes commitmenttodoso.Incontrast,Idon’tthink ratherthancommand-and-controlregulationsto I’veeverheardacandidateemphasizethatheor achieveenvironmentalobjectives. sheisrunningforofficetocontinuethepolicies TheorganizationoftheFederalReserve ofasuccessfulpredecessorofadifferentpolitical Systemfitsthisperspectiveverynicely.Members party.Politicalindependenceandnonpartisan oftheBoardofGovernorsareappointedbythe monetarypolicyprovidethepromiseofpolicy PresidentoftheUnitedStatesandconfirmedby stabilityovertime,whichinturnstabilizes theSenate.However,presidentsoftheReserve expectationsinassetmarkets.Suchstabilityand Banksareappointedbythedirectorsofthe continuityisessentialtoasuccessfulmonetary ReserveBanks,subjecttoapprovalbytheBoard policy. ofGovernors.DirectorsofReserveBankshave Centralbankindependencerequiresthatthe powersandresponsibilitiesthatareclosertothose governorhaveasubstantialtermofofficeand ofaprivatecompanythanofthoseofagovern- thatindividualpolicydecisionsnotbesubjectto mentagency.AteachReserveBank,sixofthe revisionbythegovernment.However,suchstruc- ninedirectorsareelectedbythecommercialbanks turalfeaturesofthecentralbankinstitutional thataremembersoftheReserveBank;theother designareonlythestartingpointforcentralbank threedirectorsareappointedbytheBoardof independence.Ifapresidentpubliclyattacksthe GovernorsontherecommendationoftheReserve centralbank’spolicies,thenindependencewill Bank.Thedirectorsareexplicitlynonpolitical; certainlybeincomplete.Thissubjectisavery theyaredrawnfromthelocalcommunityand 4 InstitutionsforStablePrices:HowtoDesignanOptimalCentralBankLaw arenotpermittedtoholdpartisanpoliticaloffice mercialbanks,themanagingpartnerofamajor orparticipateinpoliticalactivitythroughsuch lawfirm,CEOsofbothlargeandsmallbusinesses, activitiesasheadingcampaigncommitteesor auniversityprofessorwhoalsomanagesafamily leadingpoliticalfund-raisingefforts.Thedirec- farm,anexpertintheventurecapitalindustry, tors,inturn,selecttheBankpresidentandfirst andtheCEOofanonprofitcommunityorganiza- vicepresident,subjecttoapprovalbytheBoard tion.SomeReserveBanksincludetradeunion ofGovernors. leaders;althoughthatisnotthecasecurrently Thisinstitutionalarrangementclearlyinvolves fortheSt.LouisFed,oneoftheBank’sbranch ultimatecontroloftheFederalReserveSystem boardsdoesincludeatradeunionleader.Taking throughthepoliticalprocesscenteredonthe thetwelveFederalReserveBankstogether,direc- BoardofGovernors.Yet,aconsiderablepartof torsaredrawnfromeverysectoroftheeconomy theSystem’sleadershipobtainsofficethrough andeverygeographicregion. whatisessentiallyaprivate-sectorprocess.My EquallyimportanttotheFederalReserveis owncaseillustratesthepointnicely.Iwasauni- theflowofinformationfromReserveBankdirec- versityprofessorinRhodeIsland,withnopersonal torstoBankpresidents,whointurnusethisinfor- orinstitutionalconnectiontotheFederalReserve mationinformulatingmonetarypolicydecisions. BankofSt.Louis.IftheappointmentoftheBank Valuableinformationalsocomesfromnumerous presidentwerecontrolledbyapoliticalprocess advisorycommitteesthatmeetfromtimetotime involving,say,thestategovernorsofthestateswith attheBoardofGovernorsandtheReserveBanks, territoryintheEighthFederalReserveDistrict andfromcontactsbetweenFederalReserveoffi- (Missouri,Arkansas,Mississippi,Tennessee, cialsandtheiraudiencesastheytraveltospeak Kentucky,IndianaandIllinois),thenitisvery atvariouseventsandmeetwithbusinessand unlikelythatauniversityprofessorfromthe communityleaders.TheFederalReservehas stateofRhodeIslandwouldhavebecomeBank maintainedacontinuousassociationwithwhat president.NorisitlikelythatIwouldhavebeen areknownintheUnitedStatesas“grassroots” appointedthroughaWashingtonpoliticalprocess, contactsthroughoutthecountry.Althoughthis organizationoftheFederalReserveSystemdid giventhatIhadservedinaRepublicanadminis- notpreventthemonetarypolicymistakesthat trationbutthataDemocraticadministrationcon- contributedtotheGreatDepressionandtheGreat trolledtheWhiteHousein1998whenIwas Inflation,Ibelievethatthecurrentprocesscon- namedSt.LouisFedpresident. tributesgreatlytotheprospectsforcontinued Whatthisprivate-sectorprocessdoesisto soundmonetarypolicyintheyearsahead. reinforcethenonpoliticalnatureoftheFederal ReserveSystem.Theprocessalsoinvolvesthe ReserveBankdirectorsinanimportantway.The TRANSPARENCY FederalReservepaystheBankdirectorsverylit- tle;whattheygetoutofserviceasdirectorisan Inrecentyears,centralbankshavebecome intenseeducationinmonetarypolicy.Overtheir moreopeninmanydifferentways.Inthepast, yearsofservice,andforyearsthereafter,the centralbankersoftendiscussedmonetarypolicy directorsspreadknowledgeofmonetarypolicy inobscurewaysandseemedtorelishthemys- processesandchallengesthroughouttheircom- tiqueofcentralbanking.Asanacademic,Inever munities.Icannotimagineamoreeffectiveway thoughtthatextensivesecrecyservedcentral ofbuildingsupportforsoundmonetarypolicy bankswell,andstilldon’t. thanhavingcommunityleadersfrommanydif- Particularlygivencentralbankindependence, ferentprofessionsserveasdirectors.Consider,for opennessisessentialtopoliticalaccountability. example,thebreadthofexperienceonthecurrent Whetherbylaworconfirmedpractice,goodcen- St.Louisboard;theboardincludesCEOsofcom- tralbankdesigncallsforcentralbankstomake 5 MONETARYPOLICYANDINFLATION timelyreportsaboutpolicyactions,including First,thegovernmentshouldassignclear thereasonsforthesechanges. andobtainableobjectivestothecentralbank.I I’vediscussedtransparencyonseveralocca- favoralegislatedinflationtarget,butmoreimpor- sionsatconsiderablelength;hereIwanttomake tantthanlegislationisanunderstandinginthe twomainpoints.First,promptdisclosureofpol- societythatlowandstableinflationisthecentral icydecisionsandtherationaleforthosedeci- bank’sresponsibilityandthatthebankshould sionsisessential.However,disclosureofpolicy bejudgedonhowwellitachievesthatobjective. debatesleadinguptodecisionsmustbehandled Agovernmentmayassigntothecentralbanka extremelycarefully.Excessivedisclosurewill policygoalofcontributingtostabilityinincome damagetheopennessoftheinternaldebateand andemployment,providedthereisaclearunder- therebyincreasethelikelihoodofpolicymistakes. standingthattherecanbenocentralbanktarget Moreover,withmanydifferentviewsexpressed forthelevelofemploymentortherateofgrowth aroundthepolicytable,andviewsexpressed ofGDP. provisionallyandforthesakeofargumentand Second,thecentralbankshouldoperate thoroughness,fulldisclosureofinternaldebate independentlywithinthegovernment;thegov- withoutasubstantiallagismorelikelytocon- ernorshouldhaveareasonablylongtermofoffice fusemarketsthanenlightenthem.Ibelievethat andshouldnotbesubjecttoremovalbythe theFederalReservepracticeofdisclosingthe president,exceptforcausethroughanimpeach- transcriptofFederalOpenMarketCommittee mentprocess.Thepresidentshouldnotbeable (FOMC)meetingswithafive-yearlagworkswell. tooverturnindividualmonetarypolicydecisions Alagofthatlengthmaintainsultimateaccounta- andideallyshouldconfinecommentonthose bilityandprovidesavaluablerecordforscholars decisionstoconfidentialcommunicationswith whilepreventingdamagetothepolicyprocess. thecentralbank. Mysecondmainpointisthatpromptdisclo- Third,thecentralbankshouldbetransparent sureofpolicydecisionsandtheirrationaleis inthewayitmakesdecisionsandimplements necessaryformarketstofunctionefficiently. policy.Politicalaccountabilityrequirestrans- Monetarypolicyworksthroughmarkets;ifmar- parency;soalsodoestheefficientoperationof ketsexpectonepolicydirectionwhenthecentral themarketsthroughwhichmonetarypolicy bankintendsanother,boththemarketsandthe affectstheeconomy. centralbankarelikelytobesurprisedatsome Thesethreeprinciplesbroadlycharacterize pointanddisappointedbytheresults. allmajorcentralbankstoday.Weshouldnot, however,takethatfactasreasontoassumethat theissueissettled.Weareboundtofacestresses CONCLUSION inthefuturewhenmanywillquestionthese principles.Statingthemnow,defendingthem Thereisnouniquelyoptimalwaytowritea andexplainingthem,isourbesthopeforimprov- centralbanklawandtoinstitutionalizecentral ingpublicunderstandingandmaintainingthe bankpractices.Differentcountrieshavedifferent progressofrecentyearsthatissoevidenttoall historiesanddifferentpreferences.Letmepull centralbanksandstudentsofcentralbanking. togetherthethreadsofmyargument:Agood designforthecentralbankwillcontainthree mainelements. 6

Cite this document

APA

William Poole (2002, December 4). Speech. Speeches, Federal Reserve. https://whenthefedspeaks.com/doc/speech_20021205_poole

BibTeX

@misc{wtfs_speech_20021205_poole,
  author = {William Poole},
  title = {Speech},
  year = {2002},
  month = {Dec},
  howpublished = {Speeches, Federal Reserve},
  url = {https://whenthefedspeaks.com/doc/speech_20021205_poole},
  note = {Retrieved via When the Fed Speaks corpus}
}