speeches · February 23, 2004

Speech

William Poole · President

State of the U.S. Economy AAIMManagementAssociation St.Louis,Missouri February20,2004 I’m delighted to once again participate in becauseitallowsustosortthroughsomeofthe an AAIM program. These sessions always currenttensionsinthedatathathavespurreda help me to pull together my thinking and, lotofdiscussion.I’llthenconcludewithafew I hope, to convey some useful informa- briefremarksabouttheinflationoutlookandhow tion to you as well. theFederalOpenMarketCommittee(FOMC) About14monthsago,ifyou’llrecall,Ispoke canbestensurethatoureconomygrowsatits toAAIMonmyoutlookfortheU.S.economyin maximumsustainablerateofgrowthwithlow 2003.Atthattime,thereremainedconsiderable andstableinflation. uncertaintyaboutthelikelycourseoftheeconomy. Beforeproceeding,Iwanttoemphasizethat Theuncertaintystemmedfromseveraldevelop- theviewsIexpressherearemineanddonot mentsthathadbuffetedtheeconomyoverthe necessarilyreflectofficialpositionsoftheFederal previouscoupleofyears.Theseincludedthe ReserveSystem.Ithankmycolleaguesatthe stockmarketbust,the2001recession,theterrible FederalReserveBankofSt.Louisfortheircom- eventsof9/11,thewarinAfghanistanandpros- ments.KevinKliesenprovidedespeciallyvaluable pectofwarinIraq,risingenergypricesandsev- assistance.However,Iretainfullresponsibility eralcorporategovernancescandals.Despitethe foranyerrors. factthattheseshocksputtheeconomythrougha wringer,Iandmostofmycolleaguesthought thatthenation’seconomywasintheprocessof A LOOK BACK AT 2003 transitioningfromaperiodofrecessionandslow Inmanyrespects,2003wasafineyearfor growthtoaperiodofsolidandsustainedeco- nomicgrowth.Theeconomyenjoyedafirmfoun- theU.S.economy.Comparedto2002,economic dationbuiltuponlowandstableinflationand growthwasstrongerandinflationslightlylower. strongproductivitygrowth. Moreover,corporateprofitsrosesharplyand,in Infact,theeconomydidenjoyhealthygrowth response,thestockmarketralliedconvincingly. lastyear.Therealsurprisewasunusuallyhigh Throughitall,nominalinterestratesdeclined productivitygrowthanddisappointinglyslow modestlyandthegrowthrateoflaborproductivity employmentgrowth.Weshouldnevercomplain roseforthethirdstraightyear,registeringits aboutrobustproductivitygrowth;wecanand quickestpacesince1965. shouldcomplainthatoutputgrowthwasnotrapid IfIhadknowntheseoutcomesatthebegin- enough,giventheproductivitygrowth,toyield ningof2003,Iwouldhaveexpectedafairlybrisk robustjobgrowth. upswinginprivate-sectoremployment.Alas,as Inmyremarkstoday,Iwillreviewsomeof weallknowbynow,thatwasnotthecase,asthe lastyear’skeydevelopments.I’llthenturntoa labormarketcontinuedtoconfoundforecasters frameworkthatIfindhelpfulinthinkingabout andeconomists.Productivitygrowthwasnotjust theoutlook.Thisframework,knownasgrowth strongbutalmostoffthecharts.Asaconsequence, accounting,isespeciallyhelpfulatthisjuncture employmentgainswereminimal,astheyhave 1 ECONOMICOUTLOOK beensincethetroughoftherecessioninNovember Beginningaroundmid-year,economiccon- 2001.Whethermeasuredbythenumberofjobs, ditionsstartedtoimprove.Asmanyhadhoped asreportedintheestablishmentsurvey,orby andexpected,thepaceofbusinessfixedinvest- thenumberofpeopleworking,asreportedinthe mentbegantopickupnoticeably,asdidthepace householdsurvey,employmentdidnotkeeppace ofconsumerspending.Fortheyear,realexpen- withestimatedpopulationgrowth.I’lltalkabit dituresonequipmentandsoftwarerosenearly9 moreaboutthisdevelopmentlater.Fornow,let percent,whichwasaboutwhattheBlueChip meturntosomeofthekeydevelopmentsthat Consensushadexpected.However,contraryto occurredlastyear,andwhethertheymightsimi- expectations,businessoutlaysforstructuresfell larlyaffecteconomicactivitythisyear. forthethirdconsecutiveyear.Alsohelpingto Aboutthistimelastyear,theconsensusof boosteconomicgrowthwasasharpupturnin private-sectorforecasters,andofthePresidents exportsofgoodsandservices.Inall,withthe andGovernorsoftheFOMC,wasthatrealGDP paceofoutputgrowthrisingtomorethana6 wouldincreasebyabitmorethan3¼percentfrom percentrateoverthelasthalfoftheyear,real thefourthquarterof2002tothefourthquarterof GDPisestimatedtohaveincreased4.3percent 2003,afterincreasingabout2¾percentin2002.1 overthefourquartersof2003,noticeablyfaster Giventhefairlysharpdownturninrealbusiness thanhadbeenexpectedatthebeginningofthe fixedinvestmentbetweenthefourthquarterof year.Evenmoreimpressive,thisgainwasallin 2000andthefourthquarterof2002(12.75percent), finalsales,asinventoryinvestmentdeclined manyeconomistsbelievedthatakeyelement modestly. underpinningtheforecastfor2003wasanupturn Theroughly1percentage-pointdifference inbusinesscapitalspending.Whileadverse betweenwhattheconsensusoftheFOMCand shockshadrestrainedthepaceofbusinesspur- private-sectorforecastersexpectedfor2003and chasesofcapitalgoodssincelate2000,forecasters whatactuallyhappenedcanbetraced,inlarge agreedthattheeconomywouldcontinuetounder- part,toafewkeydevelopments.First,realcon- performunlessfirmswerewillingtocommit sumerexpendituresondurablegoodsweremuch scarceresourcestoreplacingorupgradingtheir strongerthanexpected.Inparticular,helpedby plantandequipment. generousincentives,salesofnewcarsandlight Inthefirstquarterof2003,realbusinessfixed trucksstayedabove16millionunitsforthefifth investmentcontinuedtodecline.Inhindsight,it straightyear.Second,tothesurpriseofmany, appearedthatsomeindustrieswerestillworking thehousingindustrycontinuedtopowerahead. throughexcessesthathadbuiltupovertheprior AccordingtotheBlueChipConsensusattheend fewyears.Investmentspendingwasalsocon- of2002,growthofrealresidentialfixedinvest- strainedbyanabundanceofbusinesscaution. mentwasexpectedtodeceleratefromabout7 Althoughconsumerexpendituresoverthefirst percentin2002toaround2percentin2003. halfof2003wereincreasingmodestlyfasterthan Instead,realhousingexpendituresincreasedby theywereoverthelasthalfof2002,households abitmorethan10percent,asnewhomesales werealsoexhibitinganunusualdegreeofcaution reachedarecordhighforthethirdstraightyear. overthefirsthalfoftheyear.Withexpenditures Finally,theeconomyreceivedaboostfromlarger- byconsumersandbusinessesrelativelyweak, than-expectedincreasesinrealfederaldefense realGDPgrowthremaineddisappointing,turn- expenditures. ingingainsof2.0and3.1percentatanannual Whileitispossiblethatfuturedatarevisions rateinthefirstandsecondquartersof2003, willchangethepatternofeconomicgrowthin respectively. 2003,whatmattersforourpurposestodayiswhy 1 Private-sectorforecastsfor2003aretakenfromtheDecember2002BlueChipEconometricDetail.TheFOMCprojectionsfor2003werepub- lishedintheMonetaryReporttotheCongress,whichwasreleasedonFebruary11,2003. 2 StateoftheU.S.Economy lastyear’sforecastwentastray.However,Ido supplysideoftheeconomy.Initscondensed wanttoemphasizethatthesizeoftheforecast form,thisframeworkrelatesthegrowthoflabor errorwaswellwithinnormalbounds.Although inputtothegrowthofrealGDP.Laborinputis identifyingthesizeoftheconsensusforecast analyzedbystartingwithgrowthintheworking errorisstraightforward,pinningdowntheexact agecivilianpopulation.Then,wecalculatethe reasonwhytheforecastwentofftrackisnotso percentageoftheworkingagepopulationthatis easy.Norisiteasytodeterminewhetherthe employed.Finally,weexaminethehourseach forcesthatproducedtheseerrorscanbeexpected workerputsin.Therelationshipbetweentheout- tohavesimilarinfluencesthisyear. putvariableandthelaborhoursinputvariableis Clearly,themostsignificantunexpected thegrowthoflaborproductivity—outputperhour developmentin2003wasthecontinuedstrong inthenonfarmbusinesssector.Eachyear,the growthoflaborproductivity.TheBlueChip EconomicReportofthePresidentpublishessuch Consensusprojectedthattheannualizedgrowth atable. ofoutputperhourinthenonfarmbusinesssector Thissimpleframeworkisespeciallyuseful wouldaverage2percentoverthefourquarters whenanalyzingeconomicconditionstoday,since of2003.Instead,laborproductivitygrowthaver- itremindsusthateconomicgrowthultimatelyis aged5¼percent,alargeforecasterror.Faster afunctionofemployment(oraggregatehours growthoflaborproductivitynotonlykeptreal worked)andhowproductivethoseworkersare. after-taxincomegrowthatelevatedrates,which Fromthefourthquarterof2002tofourthquarter boostedconsumerexpenditures,butithelpedto of2003,totalhoursworked—whichissimplythe keepaggregatepricepressuresatbay.Ascore numberofjobstimestheaveragenumberofhours inflationratesdriftedlowerin2003compared workedateachjob—declined0.9percent.This with2002,nominalinterestratesdidaswell. wasthethirdconsecutiveyearlydeclineinhours Besidestheobviousbenefitstothehousingsector, worked,somethingnotseensince1980-82.How- andtotheproducersofbig-ticketitemslikemotor ever,sincerealGDPcontinuedtoincrease,it vehiclesandappliances,householdsandbusi- followsthateconomicgrowthoverthisperiod nessesalsobenefitedfromlowerinterestratesby resultedfromeverfasterincreasesinlaborpro- refinancingoutstandingdebt. ductivitygrowth. Butperhapsthemostimportantinfluenceof Asanaside,oneoftheconundrumsinthe strongproductivitygrowthontheeconomyover dataoflatehasbeenthedivergencebetweenthe thepastcoupleofyearshasbeenonthedomestic twoprimarymeasuresofemployment,asreported labormarket.Letmenowturntothesecondpart onthefirstFridayofeachmonthbytheBureau ofmytalktodiscussthisimportantissue. ofLaborStatistics.Briefly,onesurvey—the householdsurvey—countsthenumberofpeople employed.Theothersurvey—theestablishment survey—countsthenumberofjobs.Thetwosur- A FRAMEWORK FOR THINKING veysdonotcomeupwiththesamecountfor ABOUT THE OUTLOOK severalreasons.One,forexample,isthatsome Whenpeopleaskmeaboutmyoutlookfor peopleholdmorethanonejobforpay,andthere- theeconomy,Iemphasizethatmyviewsare foreappearmorethanonceintheestablishment largelyinformedbylookingataconsensusof surveybutonlyonceinthehouseholdsurvey. severalforecasts.Asaninformedconsumer, Overtime,however,thetwomeasurestendto however,Idonotbuytheproductuncritically. trackeachotherfairlyclosely. Besidesexaminingtheusualforecastdetailcon- ButsinceNovember2001,establishment cerningexpectationsformajorcomponentsofGDP, employmenthasdeclinedalittlemorethan Ioftenfinditusefultoemployasimplegrowth 0.75percent,whilehouseholdemploymenthas accountingframework,whichemphasizesthe increasedalittlemorethan1.5percent.While 3 ECONOMICOUTLOOK thisdiscrepancyhasgeneratedmanymoreques- playedsuchalargeroleinuncertaintyabout tionsthananswers,bothsurveysnevertheless employmentgrowth.Usually,uncertaintyabout showthatthelabormarketremainsunusually employmentgrowthisaconsequenceofuncer- weakatthisstageofthebusinesscyclecompared taintyaboutGDPgrowth.Certainly,wearefaced withthenorm. withtheusualuncertaintyaboutGDPgrowthbut Mosteconomistsviewthepayrollestimate nowtheproductivitypuzzlesmakesanemploy- asthemorereliableofthetwoemploymentmeas- mentforecastmorethantypicallyhazardous. ures.Still,untilweseefuturedatarevisionsor Basedonpastexperience,itseemshighly satisfactoryexplanationsofthedivergenceofthe improbablethatlaborproductivitygrowthcan twoemploymentmeasuresthatresolvethe continuetooutstripthegrowthofrealGDPindefi- unusualdiscrepancy,itisappropriatetowithhold nitely,particularlywhenpopulationgrowth firmjudgmentsonsomeissues,especiallythose remainsaround1percent.Tobelieveotherwise relatedtolikelyfutureproductivitygrowthtrends. impliesfurtherdeclinesintheemployment-to- Inanyevent,itishardtoescapetheconclu- populationratio.Thatwouldinvolveagrowing sionthatin2003firmsintheaggregatewerestill labormarketdisequilibrium,whereasnormal unwillingtocompeteaggressivelyinthelabor economicforcestendtoreducesuchadisequi- marketbecausetheycontinuedtoreapimpressive libriumovertime.Ifproductivitygrowthremains productivitygainsfromtheirexistingstocksof extremelyhigh,itseemslikelythatrisingbusi- laborandcapital.Whatwedonotknowiswhether nessprofitsanddecliningunitlaborcostswill firmshadexcesslaborwhichisnowbecoming bespurstonewhiring.Outputgrowthwould morefullyutilizedorwhetherunderlyingproduc- thenrisesufficientlyaboveproductivitygrowth tivitygrowthisnowhigher.Forexample,ifFirm tobeconsistentwithsatisfactoryemployment XYZhadunderutilizedworkersitmightbeableto growth.Ifproductivitygrowthislessrapidthan produce20widgetsperworkerhourquiteeasily lastyear,thenthatoutcomewouldalsopointto whereasithadbeenproducing15perhour.How- higheremploymentgrowth. ever,ifdemandrisesfurtheritmightnotbeeasyto Thisanalysisleadsmetoexpecthigher produce25widgetsperhour,andthefirmwould employmentgrowthin2004,whichwouldlead havetoaddworkerstomeethigherdemand.On toarisingratioofemploymenttopopulation—a theotherhand,ifstructuralproductivitygrowth normalcharacteristicoftheeconomywhenreal isnowonanewhighergrowthtrack,thenhigher GDPisgrowingatahealthyclip.Itisalsoreason- demandmightbemetwiththesamenumberof abletoexpecttoseehoursworkedbegintorise, workersproducingevermorewidgetsperhour. bothbecauseemploymentrisesandbecause Datashowingthatoutputperhourroselastyear averagehoursworkedperweekrises.Judging from15to20widgetsjustdoesnotdistinguish fromthelasttwobusinessexpansions,hours betweenthesetwoexplanations. growthofsomewherebetween1and2percent Intheaggregate,then,it’snotthatproductivity seemsreasonable.Althoughprojectingproduc- growthistoohigh;theissueissimplythatreal tivitygrowthisobviouslyhazardous,aprojection GDPgrowthisnotstrongenoughtogeneratenew ofaround3percentseemsplausibletome.The jobsgiventhatproductivitygrowth.Asamatter hoursandproductivityprojectionsaddupto ofarithmetic,unlessweseerealGDPgrowthin realGDPgrowthofbetween4and5percentin excessoflaborproductivitygrowthonasustained 2004.Ithinktheseprojectionsaresensiblebest basis,wewon’tseemuchjobcreation. guesses,butIalsobelievethatthereisnoreason toruleoutthepossibilityofaconsiderablyhigher outcome.Thenormalforecastinguncertainty OUTLOOK FOR 2004 suggeststhatweneedtoconsideranerrorband Idonotrememberatimeinmyprofessional ofroughlyplusorminus1½percentfortheGDP lifewhenuncertaintyaboutproductivitygrowth forecastoverthenextfourquarters. 4 StateoftheU.S.Economy Thenetoftheseforecastsyieldssignificant fronttheFed’scredibilitywiththemarketsand increasesinemployment.TheBlueChip withthepublicishigh.Thatcredibilityisimpor- Consensusexpectsmonthlynonfarmpayroll tantforanumberofreasons.Oneofthemost employmentgainstoaveragealittlelessthan importantisthattheFederalReserve,insetting 170,000in2004.Similarly,therecentOutlook itsinterestratetargetduringtimesofuncertainty, surveybytheNationalAssociationforBusiness doesnothavetoworrythatmarketswillquickly Economicsprojectsthatmonthlyjobgainswill cometoexpecthigherinflation.Themarkets average150,000permonthin2004.Thesepro- understandtheFed’scommitmentandarepatient jectedratesofjobcreationarewellbelowthose withus.That,inmyview,istheunderlyingrea- seenduringthe1980sbusinessexpansion,when sonwhytheFOMC,initspolicystatementatthe payrollemploymentincreasesaveragedabout conclusionofitslastmeetingcouldsay,“the 230,000permonth,andduringthe1990sexpan- Committeebelievesthatitcanbepatientin sion,whenpayrollemploymentincreasesaver- removingitspolicyaccommodation.” agedroughly200,000permonth.Nevertheless, Whenviewedthroughthislens,maintaining theconsensusemploymentprojectionsfor2004 existingcoreratesofinflationaroundtheircurrent seemconsistentwithaworldofbothhigherpro- levelsmakesalotofsense.FromDecember2002 ductivitygrowthandslightlylowerpopulation toDecember2003,thechain-typepriceindexfor growthcomparedtotheprevioustwoexpansions. personalconsumptionexpenditures—thePCE Sinceemploymentgainsofroughly125,000 priceindex—increasedbyalittlelessthan1.5 permontharenecessarytokeepupwiththe1 percent,whilethecorePCEpriceindexrosebya percentannualrateofgrowthinthelaborforce, littlelessthan1percent.Keepinginflationlow theprojectedemploymentgrowthinexcessof andstabledependsimportantly,thoughcertainly notentirely,onkeepinginflationexpectationsin thatmeansthatweshouldexpectsomedecline check.Iffinancialmarkets,consumers,andpro- intheunemploymentratebytheendoftheyear ducersviewthisoutcomeasconsistentwiththe fromitscurrent5.6percentrate. Fed’slong-rungoals,thenIseenoreasonwhywe shouldnotseeasimilarlybenigninflationout- comethisyear;however,Iwouldexpecttosee THE OUTLOOK FOR INFLATION thegapbetweenthetotalandcoremeasuresnar- Besidesthereboundinstructurallaborpro- row,sinceovertimetheytendtobenumerically ductivitygrowthobservedsince1995,probably closetoeachother. themostimportantdomesticeconomicdevelop- Managinginflationexpectationsrequires mentoverthelastquartercenturyhasbeenthe appropriateresponsestoeconomicfundamentals. achievementofpricestability—or,atleast,some- Unexpecteddevelopmentshavethepotentialto thingprettyclosetoit.Today’slowandrelatively altertheFOMC’sassessmentoftheappropriate stablerateofinflationisfarremovedfromthe stanceofmonetarypolicy.Clearly,unexpected inflationexperiencefromthelate1960stothe developmentscan’tbeanticipated;iftheycould, early1980s—aperiodreferredtointheeconomic ourforecasterrorswouldconvergetozeroand text-booksasTheGreatInflation.Memoriesof theFOMCwouldonlyhavetomeetonceayear theGreatInflationarefading;manyfailtoappre- orso. ciatehowimportantthisdevelopmentis. Manyobservershavenotedthatthecurrent Inanywalkoflife,sustainedhighperform- federalfundstargetrateof1percentcannot anceaddstothecredibilityofthoseresponsible remaininforceindefinitely.Thatknowledgeis fortheoutcome.Mostpeopleunderstandthatthe builtintothetermstructureofinterestrates.The FederalReservehastheprimaryresponsibility 5-yearTreasuryrate,forexample,ishigherthan toachievethegoalofpricestability,andasacon- the1-yearratebecausethemarketexpectsrates sequenceofsustainedexcellentresultsonthat toriseovertime.Itisnotpossibletopredictthe 5 ECONOMICOUTLOOK timingofadjustmentstothefederalfundsrate target,butwearefortunatethatthemarketunder- standstheissuesowell.Iamsurethattherewill beboldheadlines,probablyonthefrontpagesof newspapers,whentheFOMCannouncesthefirst rateincrease.Suchaheadlinewillbemisleading inonesense,ratherlikeaheadlinereportingthat anairlineflightarrivedtenminutesearlyorten minuteslate.Headlinesshouldreallybereserved forunexpecteddevelopments,likeplanecrashes. Fortunately,thereisabsolutelynoreasontoantici- patemonetarypolicyheadlinesofthatsort! CONCLUDING REMARKS I’veoutlinedascenariofor2004thatappears quitepromising:ContinuedstrongrealGDP growth—perhapsevenstrongerthanlastyear’s robustgrowth;acoreinflationrateremaining around1percent,orperhapsalittlehigher;and, finally,sustainedincreasesinpayrollemployment thataresubstantiallystrongerthanthoseseen overthepastfivemonths,whichsawanaverage ofabout75,000permonth.Ifthisoutlookcomes topass,I’msurethatayearfromnowwewillall agreethat2004wasabanneryear. 6

Cite this document

APA

William Poole (2004, February 23). Speech. Speeches, Federal Reserve. https://whenthefedspeaks.com/doc/speech_20040224_poole

BibTeX

@misc{wtfs_speech_20040224_poole,
  author = {William Poole},
  title = {Speech},
  year = {2004},
  month = {Feb},
  howpublished = {Speeches, Federal Reserve},
  url = {https://whenthefedspeaks.com/doc/speech_20040224_poole},
  note = {Retrieved via When the Fed Speaks corpus}
}