1500 K Street NW, Suite 850 Washington, DC 20005 Washington Center for Equitable Growth Working paper series What’s the right minimum wage? Reframing the debate from ‘no job loss’ to a ‘minimum living wage’ David R. Howell Kea Fiedler Stephanie Luce June 2016 Working paper URL © 2016 by David R. Howell , Kea Fiedler, and Stephanie Luce. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source.
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1500 K Street NW, Suite 850 Washington, DC 20005
Washington Center forEquitable Growth
Working paper series
What’s the right minimum wage? Reframing the debate from ‘no job loss’
to a ‘minimum living wage’
David R. HowellKea Fiedler
Stephanie Luce
June 2016
Working paper URL
© 2016 by David R. Howell, Kea Fiedler, and Stephanie Luce. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source.
June7,2016
What’stheRightMinimumWage?ReframingtheDebatefrom‘NoJobLoss’toa‘MinimumLivingWage’1
DavidR.Howell,KeaFiedlerandStephanieLuce2
Nosocietycansurelybeflourishingandhappy,ofwhichthefargreaterpartofthemembersarepoorandmiserable.Itisbutequity,besides,thatthosewhofeed,cloathandlodgethewholebodyofpeople,shouldhavesuchashareoftheproduceoftheirownlabourastobethemselvestolerablywellfed,cloathedandlodged….Thewagesoflabouraretheencouragementofindustry,whichlikeeveryotherhumanquality,improvesinproportiontotheencouragementitreceives.
AdamSmith,17763
ItisanationalevilthatanyclassofHerMajesty’ssubjectsshouldreceivelessthanalivingwageinreturnfortheirutmostexertions…whereyouhavewhatwecallsweatedtrades,youhavenoorganisation,noparityofbargaining,thegoodemployerisundercutbythebad…wheretheseconditionsprevailyouhavenotaconditionofprogress,butaconditionofprogressivedegeneration.
WinstonChurchill,19064
“It seems to me to be equally plain that no business which depends for existence on paying less than living wages to its workers has any right to continue in this country…. By living wages, I mean more than a bare subsistence level — I mean the wages of a decent living.”
FranklinD.Roosevelt,19335
1WewishtothankMarkLevinsonforinvaluable,comments,adviceandoverallsupport.WearealsoextremelygratefultoDavidCooper,BobKuttner,LarryMishel,EdPaisley,BobPollin,JasonRochford,JohnSchmitt,LydiaTugendrajch,andJeannetteWicks-Limfortheircontributions,andforthefeedbackfromtheparticipantsoftheSEIUforum“MakingtheProgressiveEconomicCasefora$15Wage”(WashingtonDC,May5,2016).Ofcoursewetakefullresponsibilityfortheviewsandallthemistakes.ThispaperisanoffshootofHowell’sDecentJobsProject,generouslyfundedbytheWashingtonCenterforEquitableGrowthandtheRussellSageFoundation.2DavidHowellisProfessorofEconomicsandPublicPolicyatTheNewSchool.KeaFiedlerisadoctoralcandidateinthePublicandUrbanPolicyProgramatTheNewSchool.StephanieLuceisProfessorofLaborStudiesattheMurphyInstitute,CityUniversityofNewYork.3AdamSmith(1937),pp.68,79,81.4QuotedbyAnthonyB.Atkinson(2015),p.148.5StatementontheNationalRecoveryAct.http://docs.fdrlibrary.marist.edu/odnirast.html
1
Abstract
TheAmericandebateovertheproperlevelofthestatutoryminimumwagehasalwaysreflectedthetensionbetweenthetwingoalsofensuringdecentliving-wagejobswithmaximumjobopportunity.Themoralandefficiencyargumentsforawagefloorthatcankeepaworkerabovemeresubsistencehavealonghistory,datingbackatleasttoAdamSmith.TheU.S.federalminimumwagewasestablishedbythe1938FairLaborStandardsActtoensurea“minimumstandardoflivingnecessaryforhealth,efficiency,andgeneralwellbeingofworkers”andtodoso“withoutsubstantiallycurtailingemployment.”Inrecentyears,thebestevidencehasshownthatmoderateincreasesfromverylowwagefloorshavenodiscernibleeffectsonemployment,whichhasstrengthenedthecaseforsubstantialincreasesintheminimumwage.Buttheverystrengthofthisnewevidence—researchdesignsthateffectivelyidentifyemploymenteffectsatthelevelofindividualestablishments—hascontributedtotheadoptionofanarrowNo-Job-Loss(NJL)criterion:thatthe“right”wageflooristheonethatpreviousresearchhasdemonstratedwillposelittleornoriskoffuturejobloss,anywhere.Theeconomist’sParetoCriterion—agoodpolicyisonethatdoesnoimmediateharmtoanyone—hasreplacedtheearliermuchbroaderconcernwithaggregateemploymenteffects,andmoregenerally,withoverallnetbenefitstoworkingfamilies.Theexplicitmoralandefficiencyframingofthecaseforalivingwagebyearliergenerationsofeconomists,advocates,andpolicymakershastakenabackseattostatisticaljoustingoverwhichwagefloorwillposenoriskofjobloss(orharm)toanyone.WethinkthedebateovertheproperlevelofthestatutoryminimumwageshouldbereframedfromaNJLtoaMinimumLivingWage(MLW)standard:thelowestwageafull-timeworkerneedsforaminimallydecentstandardofliving.ThispaperillustratesandcritiquestherecentNJLframing,aswellastheusefulnessofonemetricthathasbeenheavilyrelieduponforidentifyingtheNJLthreshold—theratioofthewagefloortotheaveragewage(theKaitzindex).Wearguethattheproperframingofthedebateisnotoverthestatisticalriskofthelossofsomepoverty-wage,high-turnoverjobs,butratheroverthewagefloorthatestablishesaminimallydecentstandardoflivingfromfull-timeworkforallworkers,alongwithcomplementarypoliciesthatwouldensurethatanycostsofjoblosswouldbemorethanfullyremedied.
2
Myownviewisthatexplicitgoalsareimportant,andthatchangingthediscourseisastepontheroadtoachievingtheambition.AnthonyB.Atkinson,20156
1.IntroductionThedebateoverlowpayanditslowerboundaryhaslongrestedonmoralandeconomicefficiencyarguments.Attherootofthedebateisthehistoricalexperiencethatunregulatedlabormarketsinvariablyfailtogeneratewagessufficienttomaintainaproductiveworkforce,toensurethereproductionofthatworkforce(adequatechildrearing),andtoprovideaminimallydecentqualityoflifeaccordingtoprevailingstandards.Individuallybargainedwagesforless-skilledworkersareset,asAdamSmithputit,bythe“demandforlabour,andthepriceofthenecessariesandconveniencesoflife”(Smith1937,p.85).Butthenormalconditioninlow-skilllabormarketsisasurpluspoolofworkers(todayasin1776)which,intheabsenceofregulation,drivesthewagedownbelowefficientandmorallyacceptablelevels.Inlightofthis“hunger-discipline,”eventheAmericanneoclassicaleconomistJ.B.Clarkcalledforminimumwagelegislationasearlyas1913(Clark1913).Labormarketfailurealsoexplainstheopeningwordsofthe1938FairLaborStandardsAct(FLSA1938),whichcallsforpaythatensuresa“minimumstandardoflivingnecessaryforhealth,efficiency,andgeneralwellbeingofworkers”(FLSA1938,article202).7.Andfinally,ithelpsexplaintheriseofthelivingwagemovementinthepost-1980UnitedStatesandUnitedKingdom,alongwiththecurrent“Fightfor$15.”
Afterexperiencingsubstantialwagegainsduringtheshared-growthdecadesofthepost-warGoldenAge(1947-73),Americanworkershaveincreasinglyconfrontedlabormarketscharacterizedbyprecariousjobsthatpaytoolittletoprovideafull-timeworkerwithaminimallydecentstandardofliving.Itiswell-establishedthatAmerica’sproductivitygrowthsincethelate1970shasbeenalmostentirelyunsharedwiththevastmajorityofworkers.In2014,theaveragehourlywagesatthe10th,20thand30thpercentileswerejust$8.62,$10.08and$12.09respectively,whichisnearlyexactlywhattheyearnedininflation-adjustedtermsalmostfourdecadesagoin1979.Eventhemedianwage(the50thpercentile)increasedbyjust
6Atkinson(2015),p.140.7TheFLSAthengoesontostatethatthestandardsshouldbeimplemented“"withoutsubstantiallycurtailingemploymentorearningpower."(Article202(b)).MostoftheNJLpositionarguesfornojobslostanywhere,foranyone,whereastheFLSAtextcanbeinterpretedtorefertonetemploymenteffects.
3
85centsbetween1979and1999($16.02to$16.87),andjust3centsmoresince1999,reaching$16.90in2014(EPI2015).8Oneofthemosteffectivetoolsforensuringthatemployerspayawagesufficienttokeepalltheirfull-timeworkersabovepoverty-levelincomesisthestatutoryminimumwage.Anappropriatelydesignedlegalwagefloornotonlycanlifthouseholdswithafull-timeworkeroutofpovertybutalsoincreasetheincentivetowork,reduceswageandincomeinequality,andlessentheneedformeans-testedsocialassistanceforworkingpoorfamilies.ButthishasnotbeenthepathoftheU.S.federalminimumwage,whichhascollapsedinvaluefrom$9.54in1968to$8.00in1979toamere$7.25today(Cooperetal.2015,Table1).InresponsetoCongressionalinaction,manystatesandlocalitieshavelegislatedincreasesinthestatutoryminimumwage.CaliforniaandNewYorkpassedlargeincreasesintheirstatewideminimumwageratesinearly2016.California’swagewillberaisedinincrementsfromthecurrent$10perhouruntilitreaches$15by2022.9TheNewYorkratewillreach$15bytheendof2018foremployersinNewYorkCitywith11ormoreemployees(WoffordandTobia2016).10Even“red”(stronglyRepublican)stateshaverecentlypassedlargeminimumwageincreases.11Eightcities,includingSeattle,SanFrancisco,andLosAngeles,arescheduledtoraisethemunicipalminimumwagetoaround$15overthenextseveralyears.12Furthermore,inarecentpoll,two-thirdsofthemayorssurveyedsaidtheywouldendorsea$15minimumwage(ibid.).Theonlyrealcontroversyintoday’sminimumwagedebate,evenamongeconomists,isoverhowbigtheincreaseshouldbe.Thispaperarguesthatthereisaneedtoreframethedebateovertheappropriatetargetforthefederalstatutoryminimumwage.Areviewofthehistoricaldebatesuggeststwocontendingperspectives.Asking“Howmuchistoomuch?”,onesidesupportsahigherwage8EPI(2015)showssimilarstagnation.9Smallcompanies(25orfeweremployees)willhaveuntil2023toreachthe$15threshold.10Anadditional25citiesandcountieshavesetorraisedtheirmunicipalminimumwagesince2002,butnotallofthemhaveseta$15-per-hourwage.Forexample,Chicagosetaminimumwagethatwillreach$13perhourby2019.http://laborcenter.berkeley.edu/minimum-wage-living-wage-resources/inventory-of-us-city-and-county-minimum-wage-ordinances/11AsDavidCardandAlanKrueger(2015,p.xiii)note,“Furthermore,thefactthatcitizensinfour“red”states—Alaska,Arkansas,NebraskaandSouthDakota—votedoverwhelminglyin2014toraisetheirstates’minimumwagestoashighas$9.75anhouristestamenttothewidespreadbipartisanappealoftheminimumwageamongvoters.”12Anadditional25citiesandcountieshavesetorraisedtheirmunicipalminimumwagesince2002,butnotallhaveseta$15-per-hourwage.Forexample,Chicagosetaminimumwagethatwillreach$13perhourby2019.http://laborcenter.berkeley.edu/minimum-wage-living-wage-resources/inventory-of-us-city-and-county-minimum-wage-ordinances/
4
flooraslongasthe‘causenoharm’constraintismet.Inthisview,theproperwageissetbytheNoJobLoss(NJL)criterion:thehighestwagethatwidelyacceptedresearchhasalreadydemonstratedwillposelittleornojobloss.Ontheotherside,thequestionis“Howlittleistoolittle?”Thewagefloorisnotsetonthenegativegroundsofavoidingriskoftheunintendedconsequencesofjobloss,butratheronthepositivegroundsofensuringthatfull-timeworkcansupportaminimallyacceptablestandardofliving.Inthisview,theproperstandardistheMinimumLivingWage(MLW):thelowestwageafull-timeworkerneedstoprovideaminimallydecentstandardofliving.Thisdividecanbevividlyseeninthecurrentpresidentialelection,withHillaryClintonunwillingtosupportafederalwagefloorabove$12onaquiteexplicit“No-Job-Loss”standard,whileBernieSandershasadvocateda$15wageonstandard-of-livinggrounds.Inadditiontothisdescriptionofthefundamentaltensioninthedebate,thispaperarguesthatbecauserecentstate-of-the-artempiricalevidencehasconvincedmosteconomiststhatsubstantialhikesinthewagefloorarepossiblewithoutdiscernibleemploymenteffects,thetermsofthedebatehaveconverged,withliving-wageadvocatesoftenmakingtheircaseonNJLgrounds:amuchhigherwagefloorcanbeachievedwithoutanyjoblossandthetechnicaldebateisnowoverwheretheNJLthresholdis.Indeed,giventhegrowingacceptanceamongeconomiststhattherearemanychannelsthroughwhichwageincreasescanbeaccommodatedbeyondtheemploymentcutsrequiredbythesimple,downward-sloping-demandmodelofeconomics101,advocateshavemadethecaseforsubstantialhikesinthewagefloorontheconventionalParetocriterionofnoharmtoanyone,onthegroundsthatthehigherwagecoststhatfollowfromtheadoptionofamuchhigherminimumwagecanbeentirelyaccommodatedbyhigherproductivity,lowerturnovercosts,andhighersalesfromincreasedconsumerspending.ThispaperoffersacriticalperspectiveonthisconvergenceintheminimumwagediscoursetoanNJLframing.ThenextsectionprovidesahistoricalaccountofthechangesintherelativevalueoftheU.S.federalminimumwage,withcomparisonstothepovertyline,abasicneeds-basedbudget,themedianwage,andnationalproductivitygrowth.InSection3,weillustratetheimportanceoftheNJLcriterioninthedebate,andcritiqueit.OurcritiquedistinguishesbetweentwoNJLapproaches:thebackward-lookingNJLapproach,inwhichweareconstrainedtonavigateinthe“chartedwaters”ofstatisticalevidenceofemploymenteffectsfromwagefloorssetinotherlocationsatsomeearlierpointintime;andtheforward-looking(or“innear-real-time”)NJLapproach,inwhichincreasesintheminimumwageareimplementedintherelevantlocationandimmediatelymonitoredforemploymenteffects,which
5
isbroadlyspeakingthetakenbytheUnitedKingdominthefirstyearsoftheirNationalMinimumWage(1999-2005).InSection4,wecontendthattheheavilyreliedupon“Kaitzindex”—theratiooftheminimumwagetoanaverageormedianwage—isapoorguideforidentifyingtheNJLwagethreshold.Whileagoodmeasureoftherelativevalueoftheminimumwage,thereisnotheoreticalorempiricalreasontobelieveaparticularKaitzratiocanbeareliableindicatorofjobloss.Neitherthefactorsthatsteeremployerdecisionsonthehiringandretentionofworkersatverylowwagesnorthelaborsupplydecisionsofminimumwageworkershavemuchtodowiththemedianwageofaparticulargeographicallydefinedlabormarket(thenation,stateormetropolitanarea).Othercountries,suchasAustralia,theUnitedKingdom,andFrance,couldpushupthewagefloorbecausetheywereunconstrainedbyabackward-lookingNJLrule,whetherfixedtoaparticularwage(e.g.,$10.10)orbyaparticularpercentageofthemedianwage(e.g.,50percent).EvidenceforFranceispresentedthatsuggestsconvergencewiththeUnitedStatesoverthepasttwodecadesinemploymentperformanceforlow-skillworkers,despiteaveryhighandrisingFrenchminimumwageandanextremelylowandfallingU.S.federalminimumwage.InSection5,wesuggestthattheU.S.federalwagefloorshouldbesetbyreferencetoastandardoflivingrule—thelowestwagethatafull-timeworkerneedsforaminimallydecentlivingstandard,basedonbasic-needsbudgets.Beyondthis,wemakenospecificproposal,whichwouldbefarbeyondthescopeofthispaper.ButwedosuggestthatonepossiblemodelwouldbetosetthefederalMLWforasingleindividualinalow-modestcost-of-livingregionandcomplementitwithuniversalper-childallowances.Aquasi-governmentalbody,liketheLowPayCommissionintheUnitedKingdom,couldbechargedwithstatisticalanalysis,settingtheMLW,monitoringemploymenteffects,andrecommendingcompensatoryresponsesforanyjoblossesthatoccur.Moregenerally,agoodruleonmattersofsocialpolicyistoreturntothetaskoutlinedbyFranklinD.Roosevelt.
OurproblemistoworkoutinpracticethoselaborstandardswhichwillpermitthemaximumbutprudentemploymentofourhumanresourcestobringwithinthereachoftheaveragemanandwomanamaximumofgoodsandofservicesconducivetothefulfillmentofthepromiseofAmericanlife(PresidentRoosevelt,1937).
6
2.TheHistoricalContext:FromLoftyGoalstoaPovertyWageWhilethequestionoftheproperlevelsofsupportforthepoorspansmanycenturies,13themoderndebateoversettingalegalwagefloorappearsinboththeUnitedKingdomandtheUnitedStatesinthelate19thandearly20thcentury,justifiedonmoralandefficiencygroundsinthefaceofappallinglaborexploitation(Webb,1912;Clark,1913;Douglas,1925).ThesamemoralconvictionmotivatedtheenactmentoftheFairLaborStandardsAct(FSLA)of1938,whichestablishedtheU.S.federalminimumwage.AdvocatingforpassageoftheFLSA,PresidentRoosevelt(1937)stressedtheimportanceoffairnessinthelabormarket:“Ournationsorichlyendowedwithnaturalresourcesandwithacapableandindustrialpopulation,shouldbeabletodevisewaysandmeansofinsuringtoallourable-bodedworkingmenandwomenafairday’spayforafairday’swork.”Roosevelt’s“fairday’spay”wasdefinedasthelowestwage“necessaryforhealth,efficiency,andgeneralwell-beingofworkers,”whichtodayiswhatisusuallymeantbya“livingwage.”ThedebateoverwhatbecametheFairLaborStandardsAct(1938)focusedontheconstitutionalrightofthefederalgovernmenttointerveneinprivatevoluntarycontractsandlocalstateeconomicaffairs,ontheconsequencesforregionalcompetitivenessintheAmericansouth,aswellasoveraboutjobloss.Afteralongpoliticalstruggle,thecompromisewasanationwideminimumwagesetatjust25cents(RooseveltandPerkins’goalwas40cents).Thisamountwasequivalenttoabout$4.24in2016inflation-adjusteddollarsandcoveredonlyaboutone-fifthoftheworkforce(Grossman1978).Thefinalminimumwagepolicycontainednoformulatosetthefuturewagefloorsandnomechanismtoindexittoinflation.Accordingly,anyfutureincreaseswouldrequireanActofCongress.Thesubsequenthistoryisoneoffluctuationsaroundaverylowwagefloor.Atthesametime,theshareoftheworkforcecoveredbythefederalminimumdidincreasedramaticallyin1961andafterwardswhenthelawwasamendedtocovernewcategoriesofworkers,includingthoseemployedprimarilyinretail,localconstruction,transit,andgasstations.14Thefederalminimumwagepeakedat$9.54
13OntheexperienceintheWesternworld,seeforexample,KarlPolanyi([1944]2001)andPeterLindert(2004).14Anumberofgroupsofworkersareexemptedfromthefederalminimumwage.First,personsundertheageof20maybepaidanhourlywageof$4.25forthefirst90calendardaysofemployment.Second,employersmaypaytippedworkersaminimumof$2.13anhouraslongasthehourlywageplustipequalsatleasttheminimumwage.This$2.13tippedminimumwagehasnotbeenincreasedsince1991.(Somestateshaveincreasedthetippedwagelevel,and7stateshaveeliminatedthetipcreditaltogether.)Iftheweeklytotaloftipsplusthebasewageislessthena
7
in1968(in2014inflation-adjusteddollars).UnderPresidentReagan’sleadership,theU.S.Congressfailedtoincreasethenominalminimumwagetooffsetinflation,andtherealvalueoftheminimumwagefelltoameager$6.18in1989(Cooper2016).Figure1showsrealannualearningsforafull-timefull-yearworker(40hours,52weeks)earningtheminimumwagefrom1964to2014alongwithpovertylinesforone-,two-,andthree-personfamilies.Whilethefederalminimumwageprovidedafamilywithafull-timeworkerawagebetweenthepovertylinesofatwo-andthree-personfamilyuntilaround1982,ithassincefallentolevelsbetweenthesingle-andtwo-personpovertylines.Full-time,full-yearworkin2014wouldgenerategrosspayofonly$15,080,puttingafamilyoftwobelowthepovertyline.15Figure2offersanotherperspectiveontherelativevalueofthefederalwagefloor.Therehavebeenanumberofeffortsinrecentyearstoestimateabasicsubsistencewageforworkersindifferentfamilytypes(e.g.singleadult,singleadultwithonechild,twoadultswithtwochildren).16Mostfindthatthewageneededtopaythebasiccostsofliving—housing,food,transportation,utilities,taxes,healthcare,savings,clothing,andpersonalitems—requiresafull-timejobatawagethatis,asFigure2suggests,atleast$14forasingleperson,andsubstantiallymoreforasingleadultwithonedependentchildinlowcost-of-livingareas.Weshowthecurrentminimumwageonthisfigure,whichisabouthalfofthenecessarywagetosupportasinglepersoninsevenoftheninecities,43percentofwhatisnecessaryinBaltimoreandjustone-thirdofthenecessarywageinWashingtonDC.TheadequacyoftheminimumwagecanalsobecomparedtothetypicalpayofallworkersusingtheKaitzindex,definedastheratiooftheminimumwagetoanaverage(meanormedian)wage.AsFigure3shows,bythismeasure,theminimumwagepeakedat55percentin1968andhasdroppedprecipitouslysince,rangingfrom31-to-39percentsincethemid-1980s.
week’ssalaryattheminimumwage,employersarelegallyobligatedtomakeupthedifference,buttherehasbeenlittleenforcement.Investigating9,000restaurants,theU.S.DepartmentofLaborfoundthatin85percentofthecases,restaurantsdidnotadequatelycompensatetheiremployeesfortipincomesthatfellshortoftherequired$7.25(Cooper2016).15Foreligibleworkers(mainlywomenwithyoungchildren),asmuchasabout$2,500couldalsobereceivedfromtheEarnedIncomeTaxCredit.Ourconcernhereisnotwithtotalfamilyincome,butwiththeadequacyofearningsfromwork.16ThisincludestheEconomicPolicyInstitute’sFamilyBudgetCalculator,theSelf-SufficiencyStandarddevelopedbyDianaPearce,andtheMITLivingWageCalculatordevelopedbyAmyGlasmeier.Alloftheseprovideestimatesoftheincomeneededtocoverbasiclivingcosts,byfamilysizeandtypeaswellascityandstate.AlsoseeFredericksen(2015).
8
Stillanotherstandardbywhichtojudgetherelativevalueoftheminimumwageistosetitagainstproductivitygrowth.Figure4showsthataftermorethantwodecadesoftrackingthenation’slaborproductivity,ayawninggapbegantoappearbetweenthegrowthintheeconomyandthechangeintheminimumwage.Inshort,risingnationalwealthhasnotbeensharedwithlow-wageworkersviatheminimumwagesincethelate1960s.Thesefiguresdemonstratethatbyanyconventionalstandard,thefederalminimumwagehasfallentoextremelylowlevels.Inresponsetothisdismalperformance,throughoutthe1990sandintothe2000s,labor-communitycoalitionspressuredtheircitycouncilstoadopt“livingwage”ordinances.Theseordinancesvaried,butmostofthemappliedtofirmsreceivingcityservicecontractsandtheirsubcontractors.Somealsoappliedtofirmsreceivingeconomicdevelopmentassistance,andafewcovereddirectcityorcountyemployees.Mostordinancesdefineda“livingwage”asthehourlywageneededtobringaworkerwithafamilyoffourtothefederalpovertyline.Mostoftheseordinanceswereindexedtoriseeveryyearwiththecostofliving,andincludedadditionalprovisionsforhealthcarecoverageanddaysoffwork.Livingwageordinancespassedinover125citiesandcountiesnationwidebetween1994and2015(Luce2014).Figure1:Full-TimeEarningsattheMinimumWageandPovertyThresholdsbyFamilyType,1962—2014
Source:Author’sanalysisofEconomicPolicyInstitute(realminimumwage),andU.S.CensusBureau(2014povertythresholds)data.
Annual&Minimum&
Wage&Earnings&(52&
weeks,&40hours/week)&
Poverty&line&for&single&
person,&$12,071&
Poverty&line&for&family&of&
two,&$15,934
Poverty&line&for&family&of&
three,&$18,518
$8,000.00
$10,000.00
$12,000.00
$14,000.00
$16,000.00
$18,000.00
$20,000.00
$22,000.00
1962 1966 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 2010 2014
9
Figure2:TheMinimumWageandtheFull-timeHourlyWageRequiredforBasic-NeedsBudgetbyFamilyTypeforSelectedCitiesin2016
Source:Tungetal.(2015);ownfigure.Figure3:RatiooftheMinimumWagetotheMedianWage(Kaitzindex),1960-2014
Source:OECD.stat(dataextractedJanuary2016).
14.64 14.10 13.45 13.91 13.6215.67
14.0317.00
21.0723.59 24.06 24.90
22.67
26.76 26.40 25.86
29.58
39.35
$0.00
$4.00
$8.00
$12.00
$16.00
$20.00
$24.00
$28.00
$32.00
$36.00
$40.00
Bakersfield Phoenix Colorado=Spr Houston Minneapolis Chicago Buffalo Baltimore Washington
Single=Adult With=1=Child
Current=Federal=Minimum=Wage: $7.25
0.55
0.40
0.47
0.39
0.31
0.37
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
1960 1964 1968 1972 1976 1980 1984 1988 1992 1996 2000 2004 2008 2012
10
Figure4:TheMinimumWageandProductivityGrowth,1950-2014(1968=1)
Source:Author’sanalysisofEconomicPolicyInstitute’srealminimumwage(in2014dollars)andtotaleconomynetproductivitydata(realnetdomesticproductperhourworkedin2014chaineddollars).3.Wage-settingandEmploymentEffectsRiskofjoblossfromincreasesintheminimumwageisaclassicexampleofwhatHirschman(1991)calledthe“perversitythesis”—atleastsomeoftheintendedbeneficiariesareactuallyharmedbytheintervention.ButatleastsincetheworkofCardandKrueger(1994),ithasbecomeincreasinglyclearthattherearefewifanyemploymenteffectsthatcanbeattributedtomoderateincreasesinthestatutorywagefloor.Afterbrieflyreviewingthisevidence,thissectionconsidershowfirmssetwagesandmakeemploymentdecisions,whichbothhelpsexplainthefailureofsimpleorthodoxpredictionsofjoblossandsetsthestageforourcritiqueofthewayacommonindicatoroftherelativevalueoftheminimumwage(its“bite”)hasbeenemployedforsettingtheNo-Job-Loss(NJL)wagefloor.
a) Wage-employmenttradeoffs?ThedebateovertheproperleveloftheminimumwagehaspittedtheMinimumLivingWage(MLW)goalagainsttheeconomicinterestsofemployers,thelibertarianconcernoverthereachofthefederalgovernmentandtherightsofthestates,and,mostimportantly,theriskofjobloss.Mainstreameconomists,especiallythosetrainedintheUnitedStates,haveplayedacentralroleinthisdebate,bringingtoitanothersetofinterests—thedefenseofthebasictenetsoforthodoxeconomicdoctrine,whichisseenaschallengedbythefailureoffindevidenceofdiscernible
Real%Minimum%Wage:%Cummulative%Change
Total%economy%net%productivity:%
Cummulative%Change
0
0.5
1
1.5
2
1950 1954 1958 1962 1966 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 2010 2014
11
employmenteffects.17Itishardtootherwiseexplainthemassiveoutpouringofempiricalresearchontheminimumwage,thepublicationbias(towardshowingnegativeemploymenteffects)thathascharacterizedtheprofessionalliteratureuntilrecently,18andtheemotionalhostilityexpressedtowardfindingsthatchallengeorthodoxpredictions.19Initialevidenceontheemploymenteffectsofincreasesintheminimumwageshouldbefoundinthechangesinaggregateemploymentforat-riskworkersaroundthetimeofsudden,largeincreasesinthefederalminimumwage.Whilecrude,suchevidencewouldconfirmorthodoxpredictionsofemploymenteffectsifthe“perversityeffect”isconsequential.Indeed,theFrenchminimumwageisfrequentlyblamedforhighFrenchyouthunemployment(butseebelow).Soshouldweexpecttoobservelargeshort-runresponsesofverylargeincreasesintheU.S.minimumwageon,atleast,teenageemploymentandunemploymentrates?The1950soffersevidencethatlargeincreasesdonotnecessarilyleadtodecreasesinteenemployment.TheFLSAfailedtopegtheminimumwagetoacostoflivingindicator,soafterextendedperiodsofinflation,Congresshassometimesrespondedwithlargehikesinthewagefloor.Althoughthismayhavehadtheeffectofonlyreturningthewagetoitsformerinflation-adjustedvalue,itneverthelessconfronted
17Thecaseagainsttheminimumwageonemploymenteffectsisgroundedinsimpletextbookmodelsofthelabordemandandtheory-driveneffortstoconfirmthesepredictionshavegeneratedamassiveempiricalliterature.Atleastuntilveryrecently,foralargeshareofmainstreameconomists,thetheoreticalstakescouldhardlybemoresignificant.AstheNobelprizewinningeconomistJamesBuchananhassaid,aninverserelationshipbetweenemployerdemandforlaborandthewageis“acorepropositionofeconomics”anditsrejectionwouldbe“equivalenttoadenialthatthereisevenminimalscientificcontentineconomics”(quotedbyCardandKrueger2015,Prefacetothe20thAnniversaryEditionofMythandMeasurement).18DoucouliagosandStanley(2009,p.406)findthat“Theminimumwageeffectsliteratureiscontaminatedbypublicationselectionbias,whichweestimatetobeslightlylargerthantheaveragereportedminimumwageeffect.Oncethispublicationselectioniscorrected,littleornoevidenceofanegativeassociationbetweenminimumwagesandemploymentremains.”19TheresponsetoCardandKrueger’spapersandbookdemonstratingnoemploymenteffectswasgreetedwithprofessionalandpersonalhostility.InthesymposiumontheMythandMeasurementina1995IndustrialandLaborRelationsReviewsymposium(ILRR,July1995,vol.48no.4),FinisWelch(1995,p.848)dismissedCardandKrueger’sresearchthatwaspublishedinarguablytheworld’stopeconomicsjournal,TheAmericanEconomicReview,as“testimonytothevagariesofthereviewprocess.”Indeed,hedismissedtheentireresearchproject:“IquestionDavidCardandAlanKrueger'smodelsandhowtheydoempiricalresearch.AlthoughthenotorietysurroundingMythsuggestsimportantconclusionsthatchallengeeconomists'fundamentalassumptions,Iamconvincedthatthebook'slong-runimpactwillinsteadbetospur,bynegativeexample,amuch-neededconsiderationofstandardsweshouldinstituteforthecollection,analysis,andreleaseofprimarydata”(ibid.,p.842).ItseemsevidentthatthepasttwodecadesofresearchhaveconfirmedthevalidityofCardandKrueger’smethodsandresults.PrincetonUniversityPresshasre-issueda20thanniversaryeditionofthebook,anditremainstheclassicreference—bypositiveexample—intheminimumwageliterature.
12
low-wagefirmswithasuddennominalwageshock.The1950soffertwoexamples.OnJanuary25,1950,thewagefloorwasincreasedby87.5percent,from40centsto75cents.Thisrepresentednotjustahugeincreaseinwagecostsforlow-wageemployers,butalsoasimilarlyhugeincreaseintherelativevalueoftheminimumwage.Theratiooftheminimumwagetotheaveragehourlyearningsofnon-farmprivatesectorworkersincreasedfrom31.4percentinlate1949to56.2percentinearly1950(BLS1970,tables1.5and1.6).Whatwerethelow-wageemploymenteffects?Teenageunemploymentratesactuallyfellfrom15.8percentinOctober1949(threemonthsearlier)to15.2percentinFebruary1950(onemonthlater);theseratesfellfurthertojust12percentinApril(threemonthslater);ayearlater,inApril1951,theteenageunemploymentratewasdownto7.9percent.20Muchthesamestorycanbetoldforthe33.3percentincreaseintheminimumwagethattookplaceonMarch1,1956.21Theseepisodessuggestthat,atleastinastrongeconomy,verylargeincreasesintherealandrelativewagefloorcantakeplacewithoutobservedeffectsonjobopportunitiesforthemostvulnerableworkers.Oneofthefirstlargescaleeconometricstudiesofemploymenteffectswasreportedina1970reportbytheU.S.BureauofLaborStatistics(BLS1970),ledbyHymanKaitz(forwhomtheKaitzindexisnamed).Althoughtherehavebeendramaticimprovementsinthequalityofthedataandminimumwage-employmentresearchdesigns,thelessonsofcurrentstate-of-the-artevidence(seebelow)remainaboutthesameaswhatKaitzreportedbackin1970.
Whenallvariablesthathavealegitimateclaimtoconsiderationareincluded,themeasuresofminimumwagenotinfrequentlyhavethewrongsignand/orarenotstatisticallysignificantatconventionallevels….Ingeneral,themostimportantfactorexplainingchangesinteenageemploymentandunemploymenthasbeengeneralbusinessconditionsasmeasuredbytheadultunemploymentrate….Althoughhintsofadverseeffectsofminimumwagesshowupinavailabledata,no
20MonthlyteenunemploymentratescomefromLaborForceStatisticsfromtheCurrentPopulationSurvey(seriesLNS14000012).TheinductionofyoungmenforserviceintheKoreanWarislikelytoexplainsomeofthesedeclines,althoughthenumbersweresmalluntilmid-1950.Bytheendof1950,220,000menweredrafted,andanother552,000weredraftedin1951(https://www.sss.gov/About/History-And-Records/Induction-Statistics).21SeeHymanB.Kaitz,“ExperienceofthePast:TheNationalMinimum,”ChapterIIof“YouthUnemploymentandMinimumWages,”U.S.DepartmentofLabor,“YouthUnemploymentandMinimumWages,”Bulletin1657,1970(p.11).Thewagefloorincreasedfrom75centsto$1.00inMarch1956,whichincreasedtheratiooftheminimumwagetotheaveragehourlywagefrom43.4percentto53.2percent.Officialmonthlyteenunemploymentratesfluctuatedsubstantiallyatthistime,butthereisnoobviousupwardtrend:theMarchratewas11.5percent,aboutwhatitwasthemonthbefore(11.4percent),anditwas10.9perentinApril.Whileithit12.2percentinJune,teenunemploymentwasdownto9.8percentinSeptember.
13
firmstatementcanbemadeaboutthemagnitudeofsucheffects.(ibid.,p.11)
Whileresearchpublishedthroughoutthe1980sreportedsomenegativeemploymenteffectsforyoungworkers22morerecentandmuchmoremethodologicallysophisticatedstudieshaveshownthatminimumwagesdonotnecessarilycausejobloss.23Evenscholarswhoconcludethattheminimumwagehasnegativeemploymenteffectsgenerallyagreethatthesearedetectableonlyfordisadvantagedteenagers(Neumarketal.2014).24ThisfailuretofindrobustevidenceofnegativeemploymenteffectsofwagefloorsatthenationallevelhasbeendominatedbystudiesoftheUnitedStates,butithasalsobeenunambiguouslyconfirmedbystudiesoftheexperienceintheUnitedKingdom,whichestablishedanationalminimumwagein1999andincreaseditsharplyinrealandrelativevalueoverthenextdecade(D’arcyandCorlett2015).Moststudiesonlivingwageordinancesfindsimilarresults.25Thebenefitsofthehigherwagearesignificantforworkersbutthecostsarerelativelysmallfortheemployer.Surveysreportthatemployersareabletorecoupsomeofthecostintheformoflowerturnoverandabsenteeismandincreasedproductivity.Forexample,studiesoftheLosAngelesairportestimatethatthelivingwagereducedturnoverofbetween4percentand16percent(Fairrisetal.2005;Fairris2005).AstudyofhomecareworkerscoveredbyalivingwageincreaseinCaliforniafoundthatturnoverdecreased57percentafterthewagewasimplemented(Howes2005).Studiesforcitywideminimumwagelawsfindsimilarresults.AllegrettoandReichexaminedtheeffectsofa25percenthikeintheminimumwageonrestaurantpricesinSanJose,Californiaandfoundnonegativeemploymenteffects.Theyconclude,“theseresultsimplythatcitywideminimumwagepoliciesneednotresultinnegativeemploymenteffectsorshiftsofeconomicactivitytonearbyareas”(AllegrettoandReich2015).Prospectivestudiesoflargerwageincreasesatthestatewideornationalindustrylevelsuggestsimilarresults.Reichetal.(2016)estimatetheemploymenteffectsoftheNewYorkStatewageof$15.Theirmodelpredictsthatbysubstitutingsomeworkerswithautomation,andeliminatingsomejobsduetoproductivityincreases,22ForareviewseeBrownetal.(1981).23DavidCardandAlanB.Krueger(1995);Reichetal.(2005);Dube,etal.(2010);Allegrettoetal.(2011).ForanalternativeviewseeNeumarkandWascher(2008).24Prominentmeta-analysesoftheliteraturehavefound,onbalance,littleornonegativeeffectsonemployment(OECD2006;DoucouliagosandStanley2009;BelmanandWolfson2014).25Forareviewoflivingwageeconomicimpactresearch,seeChapmanandThompson(2006).
14
employerswouldcutapproximately41,600jobs.Inaddition,asemployerspassonsomeofthewageincreaseintheformofhigherprices,consumerdemandwoulddropsomewhat,resultinginanother36,764jobslost.Altogether,thiswouldbealossof78,364jobs.Atthesametime,thewageincreasewouldhaveindirectpositiveemploymenteffectsthroughwage-inducedincreasesinconsumerdemand.Thatisexpectedtogenerate81,532jobs—leavinganetgainof3,168jobs.Relyingon“simpleillustrativeexercisesofthephase-inofa$15wage,PollinandWicks-Lim(2015,p.1)concludethatcostincreases“couldbeabsorbedbythefast-foodindustrynotonlywithoutcausingemploymentlosses,but,crucially,withoutbusinessfirmswithinthefast-foodindustryhavingtoreducetheiraveragerateofprofitability.”
b) HowFirmsSetWagesandEmploymentHowcanemployersbemandatedtopayahigherhourlywagewithoutrespondingwithjobcuts?Therearetwoanswers.First,employerscancutormaintaintheirwagebillbycuttinghoursinsteadofworkers.Butthegeneralansweristhatemployersrarelyfaceanythingclosetoperfectproductandlabormarkets—thefoundationalassumptionofbasiclabormarkettheorythathasdominatedtextbooksforgenerations.Undertheseimperfectconditionsthereisusuallysubstantialroomforimprovingthedesignandmanagementoftheworkplace.AsJohnSchmitt(2015)hasexplained,
Someemployersmaycuthours;others,fringebenefits;stillothers,thewagesofhighlypaidworkers.Someemployersmayraiseprices(particularlyiftheircompetitorsareexperiencingsimilarcostincreasesinresponsetotheminimumwage).Someemployersmayseetheirprofitsfall(alongwiththoseoftheircompetitors),whileothersmayreorganizetheworkprocessinordertolowercosts.Someofthestrongestevidencesuggeststhatmanyemployersmayexperiencedeclinesincostlyturnover.Andworkersmayrespondtothehigherwagebyworkingharder.Anyofthesechannelsmightbesufficienttoeliminatetheneedforemploymentcutsorreducethesizeofemploymentcuts(Schmitt2015,pp.547-581).
Wewouldaddthatemploymenteffectsalsodependonwhethercostsareshiftedtohigherpaidemployeesandwhetherincreasedconsumerspendingbymorehighlypaidminimumwageworkersaffectsprofitmargins.TheResolutionFoundation(2014)comestoasimilarconclusionfortheexperienceoftheUnitedKingdomwith
15
anationalwagefloorsince1999—despiteitsrapidincreasetolevelssubstantiallyhigherthantheU.S.federalminimumwage(seefigures5and6).26Theambiguousevidenceontheemploymenteffectsoftheminimumwageisconsistentwithwhattheoryandevidencesuggestaboutwagesettinginrealworldworkplaces.Whetheranemployerwillcuthoursorworkersinresponsetoamandatedwageincreasedependsontheabilityandwillingnessofthefirmtoabsorbcostincreasesthroughproductivitygains,lowerturnovercosts,adjustmentsintheinternalfirmwage/salarystructure,orlowerprofitmargins.Theseproximatedeterminantsofthewage-employmentrelationshipareinturnareflectionofthelow-wageshareofoveralloperatingcosts,theresponsivenessofproductmarketdemandtocostincreases,andthebusinessmodelsrelieduponforcompetinginimperfectlabormarkets.27Theseproximateandunderlyingdeterminantsofthewage-employmentrelationshipwillvarysubstantiallybyestablishment,firm,region,andsector.Onehasonlytocompare,forexample,thewage-settingpracticesatWal-MartandCostco.28Botharelargediscountstoresprovidingasimilarservice,butCostcopaysitsemployeesmuchhigherwages,providesbenefits,andoffersmorehoursperworkweekthandoesWal-Mart.Totakeanotherexample,collectivebargaininghasledtolargewageincreasesfordoormenandcleanersinluxuryapartmentbuildingsintheNewYorkmetropolitanareainrecentyears,which,becauseoftheverylowlaborshareofoperatingcostsandhighinelasticityofhousingdemand,hasledtononegativeemploymenteffects.4.TheCurrentDebate:What’sWrongwith‘NoJobLoss’FramingWithlittleornoempiricalsupportfortheorthodoxpredictionofemploymenteffectsfrompreviousmodestincreasesinthewagefloorfromrelativelylowlevels,andabetterunderstandingofalternativechannelsofadjustmenttohigherwagecosts,thediscoursehasincreasinglyfocusedonwhatthemostcredibleevidencesaysisasubstantiallyhigherwagefloorthatwillrisksomejobloss.A“credible”positioninthecurrentdebaterequiresmakingthecaseforaparticularwageflooronthebasisofthisNJLthreshold,withoutreferencetohowthiswouldcomparetoawagefloorsetbyastandardoflivingwagethreshold.Thissectionbeginsby26“Researchintowhythoseofjoblosseswerenotborneoutsuggeststhatemployersadaptinavarietyofways,includingraisingprices,givingsmallerpayrisestohigher-paidworkers,reducingprofits,andboostingtheproductivityoftheirstaff”(D’arcyandCorlett2015,p.1).27Theseunderlyingdeterminantshavebeenwell-knownsinceAlfredMarshallspelledthemoutoveracenturyago.28AnothercanbefoundinClark(2014).
16
describingthecurrentoverwhelmingdominanceofNJLframingoftheminimumwagediscourseandthenturnstothecritique.a) TheNJLCriterionintheCurrentDebateandPracticeAgoodexampleoftherelianceontheNJLcriterioninmakingthecaseforalargeincreaseinthefederalminimumwageistheEPIBriefingPapertitled“WeCanAfforda$12MinimumWagein2020.”Cooperetal.(2015)makethecasethatAmericacan“afford”a$12wagein2020(worth$10.58in2014dollarsaccordingtotheauthors,or$10.92in2016dollars29)onthegroundsthatthisvalueinrealtermswasachievedbackinthelate1960s.Theauthorssupporttheircasebynotingthatthecountryisfarbetterpositionedtoaffordasubstantiallyhigherwagefloorbecauselow-wageworkereducationlevelsandtheeconomy’sproductivitylevelsarebothmuchhigherthanfourdecadesago.Forthesereasons,$12isareasonable“benchmarkfortheeconomy’sabilitytosustainaparticularwagefloor.”
Thisreportreviewsamuchwiderrangeofbenchmarksinordertoevaluatehowhighthefederalminimumwagecangoandstillfallwithinourhistoricalexperience.Anextensivebodyofresearchsincetheearly1990shasinvestigatedtheemploymentimpactsoffederal,state,andlocalminimumwagesinarangethatfallsroughlybetween$6and$10perhour.Thatresearchsuggeststhatminimumwagesinthisrangehavelittleornonegativeeffectonemployment(Cooperetal.2015,p.2).
ThispassagecontainsalltheelementsoftheNJLcriterionasdefinedabove:thegoalisthehighestwageflooralreadyestablished(withinhistoricalexperience)forwhichthereisreliableevidenceoflittleornonegativeemploymenteffects.30ConfirmingtheNJLrulelaterinthepaper,Cooperetal.writethatevidenceofwageconvergenceatthestatelevel“shouldhelptoallayconcernsthatahigherfederalminimumwagewouldhurtemploymentinlow-wagestates”(Cooperetal.2015,p.10).Cooperetal.donotaddressthequestionofwhetherahigherwage,say$15anhour,couldalsobesustainedbytheU.S.economy.OtherexamplesofexclusiverelianceontheNJLstandardincludeleadinglaboreconomistsandminimumwageresearcherswhohavestronglysupportedraisingthelegalwagefloorsubstantially,amongthemAlanKrueger,AlanManning,and29Cooper(2016),Table1.30Infact,thecurrentvalueofa2020$12wagefloor,at$10.92,isabovethe“chartedwaters”of$6-to-$10.Ontheotherhand,itmightbeviewedtobewithinhistoricalexperienceifthecriteriaweretheKaitzindex(critiquedbelow)orthelevelofproductivity(thechannelthroughwhichnationalproductivitygrowthwouldaffectthewage-employmentrelationshipforlow-wagefirmsisfarfromevident).
17
ArinDube.Krueger(2015)recentlywrotethatwhileanational$12wagefloorrisks“littleornojobloss,”a$15wagewouldtakeusinto“unchartedwaters”andthatdoingsowouldbe“arisknotworthtaking.”31Asheputit,“Althoughsomehigh-wagecitiesandstatescouldprobablyabsorba$15-an-hourminimumwagewithlittleornojobloss,itisfarfromclearthatthesamecouldbesaidforeverystate,cityandtownintheUnitedStates”(italicsadded).32Kruegerdidnotcontendthatatlevelsabove$12therewillbediscerniblejobloss,muchless“toomuch”jobloss,butonlythatsincewedon’thavetheevidence(unchartedwaters),itisnotworththerisk.Krueger’sargumentisaclearexampleofthebackward-lookingNJLstandardforsettingtheappropriateleveloftheminimumwage:anincreaseinthefederalwagefloorisnot“toomuch”ifwell-establishedevidencefromtestsonselectedjurisdictions(cities,counties,statesorforeigncountries)confirmsthatthereislittleornoriskofjoblossacrossU.S.states,citiesandtowns—anextremelystringent,andarguablyimpossible,hurdle.Aswenotebelow,thisapproachrequiresthatthelocationsthatprovidedthecrucialevidenceonemploymenteffectscouldnothaveusedabackward-lookingNJLcriterionforestablishingtheirwagefloors—ifalljurisdictionsweretorelyonanNJLrule,thehigherwagefloorsnecessaryfortestsofemploymenteffectswouldberuledoutforlackofevidence(“unchartedwaters”).Anothercriterionforsettingthewagefloorisnecessary.Inarecentdiscussionpaper,Manningmakesacompellingcasethatnegativeemploymenteffectsofmoderateminimumwagehikesare“elusive,”basedonbothareviewoftherecentstate-of-the-artevidenceandhisownevidenceforU.S.teenagersacrossstatesbetween1979and2014(Manning2016).“Evenforgroupswhereonecanestimateasizeable,robustwageeffect,theemploymenteffectishardtofind”(p.7).TheimplicationManningdrawsfromthisevidenceisthat“itisperhapstimefortheliteraturetomoveontotrytoaddressthequestionofhowhightheminimumwagecanberaisedwithoutsignificantemploymenteffectsappearing”(p.3).ThisisclearlyanexampleofNJLframing,butitisonethatisconsistentwiththepossibilityofusingaggregateemploymenteffectsasthetest(assuggestedbythe31InadebatewithBernieSanders,HillaryClintonsaid,“IdotakewhatAlanKruegersaidseriously.Heistheforemostexpertinourcountryontheminimumwage,andwhatitseffectsare.ThatiswhyIsupporta$12nationalfederalminimumwage.”http://www.ontheissues.org/2016/Hillary_Clinton_Jobs.htm32“Researchsuggeststhataminimumwagesetashighas$12anhourwilldomoregoodthanharmforlow-wageworkers,buta$15-an-hournationalminimumwagewouldputusinunchartedwaters,andriskundesirableandunintendedconsequences”(Krueger2015,p.5).Similarly,JaredBernstein,theformereconomicadvisortoVicePresidentBidenexpressedhisreservationsbyreferringtothe$15wageas“out-of-sample”:“Therecouldbequitelargesharesofworkersaffected(bya$15wage),andresearchdoesn’thavealottosayaboutthat”(NoamScheiber2015,p.A1).
18
researchdesignheusedforidentifyingwageandemploymenteffectsforteens),andnotanyjoblossanywhere—animportantdistinction.Italsoopensupthepossibilityofrelyingonforward-looking(ornear-real-time)evidenceonemploymenteffectsasthewagefloorispushedup.ThishasbeentheapproachoftheU.K.’sLowPayCommission,whichwaschargedbythegovernmentwithanNJLstandard.33
Since1999theLowPayCommissionhascommissionedover130researchprojectsthathavecoveredvariousaspectsoftheimpactoftheNationalMinimumWageontheeconomy.Inthatperiodthelowpaidhavereceivedhigherthanaveragewageincreasesbuttheresearchhas,ingeneral,foundlittleadverseeffectonaggregateemployment;therelativeemploymentsharesofthelow-payingsectors;individualemploymentorunemploymentprobabilities;orregionalemploymentorunemploymentdifferences(LowPayCommission2014,p.12).
TheResolutionFoundationsimilarlyrecommendsthatthenationalminimumwageintheUnitedKingdomshouldbesetbyaforward-lookingNJLstandard:“theLPCshouldcontinuetomaketheempiricaljudgmentofthevalueatwhichtheminimumwagecanbesetwithoutemploymenteffectsyeartoyear”(ResolutionFoundation2014,p.44).TheFoundationcallsforthetargetwagefloortobesetbythevalueoftheminimumwagerelativetotheoverallmedianwageof60percent.Thegoalisexpresslynottoachievealivingwageortoeliminatelowpay,butrathertoreduce“theUnitedKingdom’shighincidenceoflowpayfrom21percentto17percent,areasonablegoalagainstinternationalbenchmarks”(p.9)withlittleornothreatofjobloss.This17percenttargetischosenbecauseitistheOECDaverage(p.36).34“Ourview,basedonU.K.andinternationalevidence,isthatawage-floorworth60percentofthemedianwageisareasonablelodestar,indicatingthemostthataminimumwagecouldcontributetothegoalofreducinglowpayoverthemediumtolongterm(p.10).35Noreasonisgivenforwhyawagefloorsetat60percentofthe
33“OurannualremithastypicallyaskedtheLPCtoreachajudgmentonthelevelthatwillhelpasmanylow-paidworkersaspossible,withoutanysignificantnegativeeffectonemploymentortheeconomy”(LowPayCommission2016,p.vii).34This“lodestar”seemsastrangebasisforsettingtheU.K.wagefloor.TheaverageKaitzratioacrossOECDcountrieshasnoobviousconnectiontotheNJLthreshold,anappropriatelevelofbottom-endwagecompression(inequality),orminimallyacceptablestandardoflivingfortheU.K.workingfamilies.ThreeofthefiveOECDcountrieswiththelowest(best)incidenceoflowpayontheFoundation’sfigure,“LowPayintheOECD”(page37)arePortugal(7percent),Chile(9percent),andGreece(12percent),allfarbelowtheOECDaverageof17percent;fivecountrieswithhigherlow-wage-incidenceratesthantheaveragearePoland,Ireland,Israel,SouthKorea,andtheUnitedStates.35AccordingtotheResolutionFoundation,“Theminimumwagecannotsimplyberaisedto66percentofthemedianwageinordertoeradicatelowpay—eveninthelongtermthisisnot(sic)kindofroleaminimumwagecanplay”(p.44).Noreasonisgiven.Itshouldbenotedthatthis60-percent
19
medianwageistheNJLthreshold,orwhytheLPCshouldn’tgraduallyincreasethewageflooruntilsignsofintolerablelevelsofjoblossappear.Anotherleadingminimumwagescholar,ArinDube(2014),hasalsorecommendedthatthewagefloorforU.S.statesandcitiesshouldbesetrelativetothemedianwage—theKaitzratio—butunliketheLPCandtheResolutionFoundation,Dubeappearstorelyonabackward-lookingNJLstandard.Hispreferredratiois50percentofthelocalmedianwage,whichwouldbringthelegalminimumuptowhathecallsits“natural”and“appropriate”level.Dube(2014)estimatesthatiflegislatedatthestatelevel,this50percentformulawouldincreaselegalwagefloorsby26percentonaverage,reachinglevelsrangingfrom$12.45forMassachusettsto$7.97forMississippi(in2014dollars).The50percentKaitzrulewouldgeneratesubstantiallyhigherwagesthantheflat$12phased-in2020wage($10.92in2016dollars)forstateswithhighmedianwages,althoughitshouldbenotedthatmostalreadyhavehighstateminimumwages(e.g.,theMassachusettsstatewagefloorisnow$10.00).Incontrast,Dube’sproposalwouldlockinalowminimumwageanddolittletopushthelowerpartofwagedistributionuptowardthenationalaverageinstateswithlowmedianwages(suchasMississippi),whichformanyminimumwageadvocatesisoneofthemaingoalsofahigherfederalstatutorywagefloor.Aminimumwagesetat50percentofthemedianwageisnaturalandappropriate,accordingtoDube,becauseitwouldincreasethewagefloorbutposelittleornoriskofjobloss.“Overall,Ibelievethebestevidenceconcludesthatthenetimpactoftheproposedincreaseintherealstatutoryminimumwagewouldbelikelysmall,andlikelytoosmalltobemeaningfullydifferentfromzero”(ibid.p.8).Dube’sstrongestempiricalevidenceforthisclaimisthatafewstatesthathaverecentlyintroducedwagefloorsthatarejustunderthe50percentmarkhaveshownnonegativeemploymenteffects:“U.S.evidencethatsuggestssmallemploymenteffectsisbasedonanumberofstates(e.g.,Nevada,Oregon,Vermont)thathaveallraisedtheirstateminimumwagestolevelsthatsurpass46percentoftheirmedianfull-timewage”(p.9).TheexperienceoftheUnitedKingdomsince1999isalsocitedinsupportofthe50percentrule,withaKaitzIndexofaround45percent.LikeKrueger(2015)andCooperetal.(2015),Dube(2014)offersnoevidence—andindeeddoesnotcontend—thattheproposedNJLwage(inthiscase,setbythe50percentratio)marksthethresholdabovewhichthereisconsequentialandtargetisrelativetotheoverallmedian,notthefull-timemedian,whichwouldbeamuchmoreambitioustarget(andonethat,forexample,France,hasalreadyachieved(seeFigure5below).
20
escalatingjobloss.Ifthatthresholdactuallyturnsouttobe55percent,65percentor75percent,therewouldbehugeforegonewagebenefitsforlow-wageworkersandtheircommunities,andfortheeconomyasawhole—atnocostinjobs.Thesame,ofcourse,goesforsettingthetargetat$12,withoutevidencethat$13.50,$15,or$17istheNJLthreshold.Insum,thebackward-lookingNJLruleisthereforeanecessarilyconservativecriterionbyconstruction—onethatmustbelowerthantheNJLthresholdbutbyhowmuchisunknown—potentiallyleavingagreatdealof“wagesonthetable.”b)What’sWrongwiththe‘NoJobLoss’CriterionWebrieflyoutlinesixproblemsposedbyrelyingontheNJLcriterionforsettingtheappropriatewagefloor.
1) TheStatisticalProblem:theLimitsofaPurelyStatisticalCriterion
IdentifyingtheNJLwagefloorfromeconometricevidenceisbothextremelychallengingandinherentlycontroversial.Giventhemanypartieswithbigstakesintheoutcome,relyingonastatisticallyderivedNo-Job-Lossthresholdallbutguaranteesendlessdebatesoverempiricalresearchdesign.36Forexample,Cooperetal.(2015)pointoutthattheevidencewehaveonemploymenteffectsisbasedonincreasesintheminimumwagewithinthe$6-to-$10range(althoughmostofthosetestswereinfactconductedwithreferencetowagefloorsbelow$9.00).AlthoughSchmitt(2013),Manning(2016),andmanyothershaveconvincingly(tous)concludedthatthebalanceofevidenceisthatdiscernibleemploymenteffectsofmodestincreasesinlegalwagefloorsremainelusive,manyresearcherscontinuetoargue,withcrediblestatisticalsupport,thatsizableincreases,evenjusttolevelsatthetopofthis$6-to-$10range,willcauseatleastsomejoblossinsomeestablishmentssomewhere.RequiringthatthereisevidenceofNJLtothesatisfactionofmostresearchersandinterestedpartiesleavesthesettingoftheminimumwagesubjecttotheoutcomeofastatisticalcontestoverwhosedataandresearchdesignsaremostcredible—adebateinwhichhardlyanyonecanmeaningfullyparticipateexcepttheresearchersthemselves—seemsapoorbasisformakingpolicythatmatters.
36Thismaybeparticularlytruegiventhenatureofsomeofthelow-wagesectors,wherethereisusuallyveryhighturnoveramongemployees,andahighfailurerateinsomeindustries.Forexample,onestudyfoundthatalmost60percentofallrestaurantscloseorchangeownershipwithintheirfirstthreeyears(Parsaetal.2005).
21
2)TheMethodologicalProblem:HowareHigherWageFloorsSetintheFirstPlace?Asthehighestwagefloorpossiblethatposeslittleornoriskofjobloss,theNJLthresholdrequiresreal-worldevidencethatcanidentifythewagefloorabovewhichdiscerniblejoblossislikely.Butsinceitisevidence-based,withinanygivenjurisdiction(say,attheleveloftheUnitedStates),thiscriterionrulesoutthepossibilityofsettingthewagefloorsinthefirstplace.Wagefloorsmustprecedetheempiricaltestsoftheireffects.Inshort,thebackward-lookingNJLrulerequiresthatthereareplacesthatdonotsetthewagefloorwiththiscriteriontogettheevidenceforsettingawagefloorthatsatisfiestheNJLcriterion—likeSeattle,orCaliforniaandNewYorkState,orFrance,Australia,andtheUnitedKingdom.Butthatmeansthattheevidencemustcomefromlocationsthatuseothergroundsforsettingthewagefloor,likestandard-of-livingcriteria.This,inturn,meansNJLjurisdictionsarecondemnedto“followingtheleader,”oralwaysbeingyearsbehindothercountries(orlocaljurisdictions)whilewaitingforevidencethatraisingthewagefloorgenerateslittleornojobloss.Extrapolatingfromtheexperiencesofotherjurisdictionssomeyearspreviouslyraisesotherproblems.Wemustassumethatthestatisticaleffectsofawagefloorinlabormarketsinwhich,forexample,thereissupportforaMinimumLivingWage(MLW)criterion(sayAustraliaorFranceorSeattle)applytoalocationwhosesocialnormscallonlyforanNJLrule.Anevidence-basedapproachwouldneedstatisticalsupportforthisexpectation.Moregenerally,theremaybemanyotherreasonsfordoubtingwhetherolderevidencefromMLWjurisdictionscanbeexpectedtopredicteffectsofafederalminimumwageacrossthenation’sstates,counties,cities,andtowns.ItshouldbeclearthattheNJLcriterion,especiallyonethatis“backwardlooking,”cannotstandonitsownasacoherentandmeaningfulstandardforsettingthelegalwagefloor.3)The“MoneyLeftontheTable”ProblemCloselyrelated,itistellingthatneitherCooperetal.(2015),Krueger(2015),Dube(2014),orManning(2016),theResolutionFoundation(2014),northeLowPayCommission(2014,2016)citeevidencethatcrediblyidentifiestheNJLthreshold—thewagebelowwhichitisknownthatthereislittleornoriskofjoblossanywhere,andabovewhichthereisknowntobeariskofjoblosshighenoughtobe“notworthtaking,”asKruegerputit.Adherencetoabackward-lookingNJLcriterionprecludessettingawagefloorhighenoughtodiscoverthisthresholdbydefinition—thereisnoevidenceunlessahigherthanexistingwagefloorisset,andthisrequiressome
22
othercriterion.Asaresult,theNJLapproachtosettingtheminimumwagemustleave“moneyonthetable,”meaningthewagefloormustbebelowtheNJLthreshold.37
4)TheNetMonetaryBenefitsProblem:WhyisJobLosstheOnlyConsideration?TheNJLcriterionisconcernedonlywiththecostsideofanincreaseintheminimumwage.Framingthequestionthiswayensuresthatthepossibilityofcostsrelatedtojoblossnecessarilytrumpsthecertaintyofthebenefitsofwagegains,bothdirectlyfortheworkersreceivingwageincreasesandindirectlybyreducingtheneedforsocialspendinginsupportofworkingpoorfamilies.Itiswell-establishedthatiftherearenegativeemploymenteffects,theyaresmall,reflectingasteep(highlyinelastic)labordemandcurveevenfordisadvantagedteens(CardandKrueger2015;Manning2016),whichmeansthatawagehikeincreasesthetotalwagestoworkers.TheNJLstandardforsettingtheappropriatewagefloorentirelyignoresthemaintraditionaljustificationfortheminimumwage—themoral,social,economicandpoliticalbenefitsofamuchhigherstandardoflivingfromworkforlowincomeworkersandtheirfamilies.
5)TheEquityandEfficiencyProblems:WhyShouldn’tFairnessandSystem-WideWelfareGainsCount?
TheNJLcriterionfailstotakeintoaccountthefundamentalethicalandefficiencyjustificationsthathavemotivatedlivingwageadvocatessinceatleastthe19thcentury.Thesejustificationscanbeusefullyexpressedintheformofthreearguments:sustainability,capability,andexternality(Stabile2008;WernerandLim2015).Onbothhumandignityandeconomicefficiencygrounds,workersshouldbeabletosustainthemselvesandtohavetheopportunitytodosofromemploymentratherthanfromeithertax-basedredistributionorprivatecharity.Thecapabilityargumentextendsthatofsustainability.AsRyan(1912,p.73)argued,thewageshouldbehighenoughtomeetthe“minimumconditionsofrightandreasonableliving,”whichisnecessarytoa“lifeworthyofahumanbeing”(WernerandLim2015,p.5).Theaffronttohumandignitydimensionisbestexpressedinthewordsofworkersthemselves.Thefollowingisanonymous,from1935.
37Aforward-lookingNJLrule,byallowingaregularincreaseintherealandrelativewageflooruntilthereisactualevidenceofjobloss,couldproducethisthreshold,andbyknowingit,wecouldthencalculatethetheforegonewagesattributabletobackward-lookingNJLframing.
23
Themostsurprisingdayeverseeninthisplacewasyesterdaywhenthebosswasorderedtopayusthecoderate…Youcanguessthemoneyishandy.Butthereissomethingmorethanthemoney.Thereisknowingthattheworkingmandon’tstandaloneagainstthebossesandtheirsmartlawyersandalltheirtricks.Thereisagovernmentnowthatcareswhetherthingsisfairforus.Itellyouthatismorethanmoney.Itgivesyouagoodfeelinginsteadofallthetimeburningupbecausenothingisfair(Waltman2004,p.183).
FollowingSen(1999)andStabile(2008),WernerandLim(2015)addamoreinstrumentalargument,onethattiesmoralitytoefficiency:adecentwageisnecessaryfor“people’scapabilitiesinregardtothefunctionstheyhaveasmembersofsocietyandasworkers,andtotheirabilitytoenhancecapabilitiesintheirchildren.”Althoughoftenneglectedinthecurrentdebate(butwascommoninthelivingwagediscussionacenturyago),thereisalsoapowerfulefficiencycaseforaMinimumLivingWage(MLW)onexternalitygrounds.Awagethatfailstomeetminimalthresholdsforsustainabilityandcapabilitymeansthatemployersareimposingsocialcostsonthenation,whichiswhySidneyWebbreferredtopaymentofbelowsubsistencewagesasa“viciousformofparasitism”(Webb,1912;seealsothequotebyRooseveltonthetitlepage).UsinglanguagelikeSidneyandBeatriceWebb,theU.S.SupremeCourtstatedthefollowingintheirmajorityopinioninWestCoastHotelv.Parrish(1937)that:
Denialofalivingwageisnotonlydetrimentaltotheirhealthandwellbeing,butcastsadirectburdenfortheirsupportonthecommunity.Whattheseworkersloseinwagesthetaxpayersarecalledupontopay.Thebarecostoflivingmustbemet.….Thecommunityisnotboundtoprovidewhatisineffectasubsidyforunconscionableemployers(QuotedinAnker2011,p.78).
Inadditiontoadrainongovernmentresources,employerspayinglowwagescoulddrivedownwagesinotherindustries.ACongressionalResearchServicereportontheFLSAnotes:
Atthetimeoftheact’spassage,Congressfoundthatafewemployerswhopaidsubstandardwagescausedadecreaseinwageswithintheirrespectiveindustries,becauseotheremployerssoughttocompeteinthemarketplacewithlowerpricedgoods.Congressalsofoundthatthesedecreasedwagescausedone-thirdoftheU.S.populationtobe
24
“ill-nourished,ill-clad,andill-housed(CongressionalResearchService2013,p.1).
ThecaseforrelianceonaNJLcriterionwouldhavetorestonthebeliefthatthecostsofjoblosswouldmorethanoffsettheethicalandefficiencybenefitsofahigherwagefloor.Thisisthedebatethatisneeded.6)ThePublicPolicyStraitjacketProblem:ASpecialHurdleforLow-WageWorkers?
ThefinalproblemconcernstheimplicationsoftheNJLcriterionforpolicymaking.Requiringthatanewpolicymustpassthishurdleasestablishedbytheagreementamong“experts”regardingthestatisticalevidencewouldeffectivelyruleoutmanynewpoliciesandregulations.Fromahistoricalperspective,strictadherencetosuchacriterionforpolicymakingwouldhavemadeitimpossibletopassahostofregulationsandpoliciesthathaveemploymenteffects,suchaschildlaborlaws,occupationalhealthandsafetyregulations,andenvironmentalregulations.Insum,muchofthepolicydebateovertheappropriateleveloftheminimumwagehasbeenconductedasiftheonlyrealquestionisoverthewagefloorthatposeslittleornoriskofjobloss.Butasamatterofstatisticalpractice,thisisnearlyimpossibletoestablish;itfailsmethodologicallyasageneralcriterionforsettingwagefloorsbecausegeneratingtherequiredevidencerequiresthathigherthanexistingwagefloorsmustbeset,whichinturnrequiressomeothercriterion;itisbyconstructionconservative,likelytoleavemuch“moneyonthetable”;itentirelyignoresthenetbenefitsofawagethatmaycausesomejoblossbutiscertaintoproducefarlargerbenefitsintheformofhigherwageincomes;itignoresimportantequityandefficiencyconsiderations;anditrepresentsanextraordinarilyconservativeruleforpolicymakingthatiftakenseriouslyoverthecourseofthelastcenturywouldruleoutmuchofthesocialregulationthatcurrentlygovernsU.S.society.5.IdentifyingtheNJLThreshold:WhatRolefortheMedianWage?Whileaforward-lookingNJLrulewouldnotgenerateawagefloorthatmaximizesnetmonetarybenefitstolow-wageworkers,muchlessoverallbenefits(accountingethicalandefficiencyconsiderations),itwouldatleastoffertheprospectofidentifyingtheNJLthreshold.Forthispurpose,itwouldbeusefultohaveasimplemetricthatcouldbeusedtopredictjoblossinresponsetoaminimumwageincrease.ManyhavepointedtotheKaitzIndex—theratiooftheminimumwageto
25
themedian(orsomeotheraverage)—asjustsuchametric.Itisausefulmeasureoftherelativegenerosityand“bite”oftheminimumwage:thecloserthewageflooristothemedian,themorecompressedthebottomofthewagedistribution,andthelargerwillbethenumber(andshare)ofworkersaffected.Inthedebateovertheminimumwage,ithasbeenashortstepfromtheuseofthisratiotomeasuretherelative“bite”oftheminimumwage,toitsuseasareliableguidetotheriskofjobloss.TheheavyrelianceontheKaitzindexintheCooperetal.andDubepapers,andbytheResolutionFoundationandLowPayCommissionintheUnitedKingdom,illustratethegreatappealofthissimplemetric.ButisitameaningfulguideforsettingtheNJLwagefloor?Wethinknot.Asnotedabove,thedeterminantsofwage-setting,andconsequentlythedynamicsofthewage-employmentrelationship,arecomplexandvaryacrossestablishments,firms,industrysectors,andregions.Itisnotjustthechangeinthelaborshareofoperatingcoststhatmatters(whichwillbedeterminednotjustbythewageincreasesbutalsobyrelatedchangesinproductivityandturnovercosts),buttheabilitytoaccommodatethesechanges,whichwilldependontheelasticityofproductdemand,changesinconsumerdemandandemployerprofitability(andwhatisviewedbyemployerstobean“acceptable”levelofprofitability),andallthesefactorswillallbeimpactedbyprevailingsocialnorms.Themedianwage,definedbysomegeographicarea,notonlyfailstocapturethesecriticaldimensionsofwageandemploymentsetting,butisalsoinherentlyproblematicasausefulguidebecauseitwillvarywiththejurisdictionallevel(reflectingtherelevantgeographicarea’smedianwage),whichislikelytobelargelyindependentofthefactorsthatdeterminetheemploymenteffectsofamandatedwageforverylow-wageworkers.Ahelpfultheoreticaljustificationforthemedianwageasaguidetothe(presumed)tradeoffbetweenwagesandemploymentwouldneedtoexplainwhichreferencewage(overallmedianormean?forallworkersorjustfull-timeworkers?forhourlywageworkers?forteenageworkers?)andwhichjurisdictionallevel(nation,state,county,city,town,neighborhood?)wouldbetherightonesforidentifyingboththeNJLthresholdandthelikelyemploymenteffectsofwagefloorsabovethethreshold.a) TheKaitzIndexandtheLow-WageEmploymentDecision
Weknowofnosuchatheoreticaljustificationintheliterature,butDube’s“Proposal13:DesigningThoughtfulMinimumWagePolicyattheStateandLocalLevels”(Dube2014)offerssomearguments.HegivesthefollowingexplanationfortheusefulnessoftheKaitzindexasaguideforsettingthestatutorywagefloorandforwhya50-percentratioisthemostappropriate.
26
Therearethreereasonstopayattentiontothismeasure,especiallyusingthemedianasthereferencewage.First,acomparisonoftheminimumwagetothemedianoffersaguideforhowbindingaparticularminimumwageincreaseislikelytobe,andwhattypeofwagethelabormarketcanbear…ahighratio—sayaround.8—indicatesahighlyinterventionistpolicywheretheminimumwageisdramaticallycompressingdifferencesinwagesfornearlyhalftheworkforce.Second,thiscomparisonalsoprovidesuswithanaturalbenchmarkforjudginghowhighorlowaminimumwageisacrosstimeperiodsoracrosscountriesthatvaryintermsoftheirlabormarketsandwagedistributions.Third,themedianwagealsoprovidesanaturalreferencepointforjudgingwhatisareasonableminimumwagelevels:nooneexpectsthattheminimumwageshouldbesetequaltothemedianwage,butfairnessmaybecomeafactorwhentheminimumwagefallsbelow,say,one-fourthorone-fifthofthemedianwage(Dube2011,p.2).
Reason#1ispresumablythemostimportant,anditiscertainlytruethattheratio“offersaguideforhowbindingaparticularminimumwageislikelytobe.”Butthisis(almost)truebydefinition—ahigherminimumwagewilltendtoaffectmoreworkers—andshortofagoodtheoryorcompellingempiricalevidence,thisalonetellsuslittleabout“whattypeofwagethelabormarketcanbear.”Dubereferstotwoextremeexamples,butnoreasonisgivenforwhyweshouldbeconfidentthata20-percentKaitzratiowillnotcausediscerniblejobloss,whilean80-percentratiowillsurelygenerateunacceptablylargejoblosses,orforthatmatter,why50percentistherightratio.Reason#2istheclaimthattheKaitzindexisausefulmetricforcomparisonsoftherelativevalueofthewageflooracrossjurisdictionsandovertime.Thisiscertainlytrue,butitisnotareasonforrelyingonthisindicatorasagoodguidetotheriskofjobloss,muchlessasthemeansforidentifyingaparticularNJLwagethreshold.Reason#3appealstowhatis“natural,”“reasonable,”and“fair,”butdoesnottelluswhatmakesaparticularratenatural,reasonableandfair.Isthepreferred50-percentrationaturalandreasonablebecauseitreflectsafair(equitable)degreeofwagecompression(bottom-endwageinequality)orbecauseitistheratioatwhichtherewillbenojobloss,orboth?HowdoweknowthatwhatisviewedtobefairandwhatturnsouttobetheNJLthresholdwillbethesameacrossjurisdictionsasdifferentastheUnitedStatesandFrance,MississippiandMassachusetts,ormetropolitanMiamiandSanFrancisco?
27
Todemonstratewhathis50-percentproposalwouldmeanatthelocallevel,Dubepresentstablesthatshowthe“targetminimumwage”generatedbya50-percentKaitzratioforstatesandselectedmetropolitanareas.Totakejustoneexample,accordingtoDube’sTable13.1,thewagefloorshouldbe$12.25fortheNewYork-Newark-JerseyCitymetropolitanarea.Butifthemedianisagoodbenchmarkonequityandjoblossgrounds,thenNewark’stargetwagefloorshouldbefarlowerthanManhattan’s:becauseNewark’smedianwageismuchlowerthanManhattan’s,a50-percentKaitzratiowouldrequireamuchlowerlegalwagefloorforNewark.Inshort,implementingthe50-percentKaitzruleattheleveloftheNewYorkmetropolitanareamustwillleadtojoblossinNewJerseyifthisratioisagoodapproximationfortheNJLthreshold.Onthissameassumption,usingauniformKaitzratiotosetwagefloorsatthestatelevelwouldlikelyposeevengreaterthreatsoflocaljobloss.Ataminimum,thisexamplesuggeststhattheKaitzratiowouldbeexpectedtodoabetterjobasapredictorofjoblossthesmallerthejurisdictionalunit.Butifthat’srightthenwhatmakesNewarktherightgeographicunit?Toextendtheexample,wecanbesurethatNewark’smedianwagewillbeaffectedbychangesinthecompositionofNewark’semploymentbase.Let’ssaythatabigincreasetakesplaceintheshareofbigheathcarefacilities,nonprofitorganizations,andgovernmentagenciesthathaverelativelyhighwages.Withthisgrowth,thecity’scity’smedianwagewillincrease,causingtheratiooftheminimumwagetothemedianwagetofallandconsequentlysowillitsKaitzratio—let’ssayfromtheoriginal50percenttothe37percent,the2014figurefortheentirecountry.Tomaintainthe50-percentrule,Newark’swagefloorwouldhavetobeincreased.Butthereisnoobviousreasonwhythesecitywideshiftsinemploymentcompositionshouldmakeitpossibleforinnercityretailshopsandfastfoodrestaurantsthatcatertoneighborhoodcustomerscaneasilyadjusttothenew,higherminimumwage.Inthisscenario,the50-percentKaitzratioshouldbeappliedtosubsectionsofNewark,leadingtodifferentwagefloorswithinthecity’sboundaries.ThisillustratesthepotentialimpracticalityofusingaparticularKaitzratiotoachieveanNJLwagefloor.38Inshort,themedianwageseemsapoorguidetosettinganappropriatelegalwagefloor—onethatishighenoughtomakeadifferencebutnotonethatrisksjobloss.A
38IftheKaitzindexprovestobebetterperformingasaguidetojoblossatsmallerjurisdictionallevelsthenthechallengethenbecomesthefeasibilityofupdatingthelocalKaitzratioandthereforethelocalminimumwageasthereferencemedianwagechanges.Thereisalsotheobviousinefficiencyandunfairnessofstatutorywagefloorsthatvaryacrossneighborhoods,fromcitytocity,andacrossstatesthatsharethesamemetropolitanarea,asinthecaseofmetropolitanNewYork.
28
thought-experimentthatextendsthisdiscussiononthedifficultiesposedbyageographicallydefinedKaitzindexforsettingthewagefloorappearsintheAppendix.b) ImplicationsoftheCross-CountryEvidence
Withouttheory,orevenstrongintuition,forwhyaparticularKaitzratiowoulddoagoodjobofidentifyinganNJLwagefloor,itsusefulnessisanempiricalquestion.Wewillfocusonthecross-countryevidencefromtheOECD,whichillustrateshowwidelyvaryingtheKaitzratiois,bothovertimeandacrosscountries.FocusingmainlyonacomparisonoftheUnitedStates,France,andAustralia,weshowthatthereislittleevidenceofalinkbetweentheKaitzindexandemploymentperformance.Thatis,countriesthathavechosentosetthelegalwageflooratahighlevelrelativetothemedianwage(FranceandAustralia)donotshowsystematicallyhigherunemploymentrates,orloweremploymentrates,thantheUnitedStates.Nordoweobservethepredicteddivergenceovertime:astheKaitzratiohasrisentoover60percentforFranceandAustraliawhilefallingtoaround37percentfortheUnitedStates,thepredictedwideninggapinemploymentperformancedoesnotappearinthedata.Wealsocomparetheincidenceoflowpayandemploymentperformanceacrosscountries.Asthelegalwagefloorincreasesrelativetothemedian—ahigherKaitzindex—thebottomofthewagedistributioniscompressed,leadingtolowerbottom-endinequality(wagedispersion)aswellasalowerincidenceoflowpay.Theconventionaldefinitionofthelowpaythresholdis2/3ofthemedianwage—theconventionaldefinitionoflowpay.AhigherKaitzratiowillreducetheincidenceoflowpayastheminimumwageconvergestothelow-paythreshold.Wealsofindnocorrespondencebetweentheincidenceoflowpayandtheemploymentrateofyounglesseducatedworkers.Figure5showsthesameseriesalongwithKaitzratiosfortheUnitedStates(Figure3)andfourotherrichcountries.Sincethemid-1980stheUnitedStateshasheldthelowestposition.By2014,theU.S.ratioofthewagefloortothemedianwagewasonly37percentcomparedto61percentforFrance,53percentforAustralia,48percentfortheUnitedKingdom,and45percentforCanada.ThisfigureshowsthatwhileFrancemaintaineditscommitmenttoahighandrisingwagefloor,theUnitedStatestooktheoppositepath.
29
Figure5:TheRatiooftheMinimumWagetotheMedianWage(“KaitzRatio”)forFiveCountries,1960-2014
Source:OECD.stat(dataextractedJanuary2016)AsimilarpatternisshowninFigure6fortheminimumwageintermsoftheabsolutevalueoftheminimumwage,withFigure6areportingpurchasingpower(usingtheOECD’sindexofpurchasingpowerparity)andFigure6bshowingthevalueintermsofexchangerates.BothmethodsofvaluingtheminimumwageputAustraliaandFrancefarabovetheothercountries,withtheUnitedStatesatthebottom.TheperformanceofAustraliaisparticularlynotable,inthatitillustratesthesensitivityoftheKaitzindextochangesinthevalueofnotjusttheminimumwage(thenumerator)butthemedianwage(thedenominator).Figure5reportsafairlystrongdeclineintheAustralianKaitzratio,fromwellabovetowellbelowthatofFrance.ButasFigures6aand6bshow,thebuyingpowerofAustralia’sminimumwage(howevermeasured)hascontinuedtoincrease;thereasonforthedeclineintheKaitzindexshowninFigure5isnotbecausethegovernmentfailedtoraisethevalueofthewagefloor(perhapsforfearofjobloss?),butbecausethemedianwageincreasedevenfaster.ThisunderscoresthepotentialdifficultyofrelyingonafixedKaitzratioastheNJLtarget,asproposedbyDube(2014),theU.K.government(LowPayCouncil2016)andtheResolutionFoundation(2014).
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
2014
France
Australia
United:Kingdom
Canada
United:States
30
Figure6:TheValueoftheNationalMinimumWagesforFiveCountriesinConstantUSDollars(2014),1960-2014a.Measuredin2014constantpricesusingUSDollarPurchasingPowerParities(OECD)
b.Measuredin2014constantpricesusing2014USDexchangerates(OECD)
Source:OECD.stat(dataextractedMay2016)
Australia
Canada
France
UK
US
$2
$4
$6
$8
$10
$12
1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015
Australia
Canada
France
UK
US
$2
$4
$6
$8
$10
$12
$14
$16
1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015
31
FranceandAustraliaofferagoodexamplesofnationalcommitmenttoaMinimumLivingWage.InthecaseofFrance,theminimumwagewasestablishedanddesignedtopromotesocialinclusionthroughemployment.AsCaroliandGautie(2008,p.18)explain,theFrenchminimumwagereflects“apoliticalandsocialconsensusaccordingtowhichadecentwageshouldbedefinednotonlyinabsolutebutalsoinrelativetermsandthuslowwagesshouldbenefitfromgrowthandbeindexedtotheaveragewageincrease.”Likewise,accordingtotheWorkplaceRelationsActof1996,thegoaloftheAustralianminimumwageistoensurea“fair”wage,judgedaccordingtolivingstandards“generallyprevailingintheAustraliancommunity”(Healy2011,p.633).
Accordingtotheconventionalwisdom,Francehaspaidaheavypriceforthiscommitmenttoeradicatingextremelowpay,intheformofalong-runworseningofemploymentoutcomesforyoung,less-educatedworkers,especiallyrelativetocountrieswithacommitmenttoflexible,lessregulatedlabormarketssuchastheUnitedStates.Butthedatasuggestotherwise.Torecap,whilebothcountrieshadaKaitzindexofaround45percentto47percentinthelate1970s,bythemid-2000s,theFrenchratiohadrisenabove60percentandtheU.S.ratiohadfallento31percent.TheU.S.-Francegapwasequallymassiveinrealpurchasingvalues:astheU.S.minimumwagefellfromaround$9.50to$7.25,theFrenchminimumwagerosefrom$7.00to$11.00(in2014dollars-Figure6a).Hastheopeningofa25percentagepointgapintheKaitzindexforFranceandtheUnitedStates—onethatisreinforcedbyabuyingpowergapthathasreached$3.75—ledtoadivergenceinemploymentandunemploymentratesforthemostvulnerableworkersinthesetwocountries?Figure7ashowsthattheconventionallydefinedunemploymentrateforyoungworkers(ages15to24)wasmuchhigherinFrancethanintheUnitedStatesinthemid-1980sandmid-1990s(theheightoftheEuropeanrecession).However,thisgapnarroweddramaticallyinthe2000s,andespeciallyduringthe2008-10financialcrisis,whiletherelativevaluesoftheFrenchandU.S.minimumwagescontinuedtosharplydiverge.Withtheexceptionof2011to2013(inpartareflectionofthecommitmenttoEuropeaneconomicausteritypolicies)thereisnoevidenceofasecularwideninggap,aswouldbeexpectediftheKaitzindexwasagoodpredictorofemploymentperformanceforthemostvulnerableworkers.Figure7:AlternativeUnemploymentRates(U/LFandU/POP)forAges15-24,1983-2014*
32
a.TheUSandFrance
b.TheUSandAustralia
*U/LFisthestandardunemploymentrateandistheratiooftheunemployedtothelaborforcefor15-24yearolds;U/POPistheratiooftheunemployedtothepopulationfortheseages.Source:OECD.stat(dataextractedJanuary2016).Figure7aalsoshowsthattheorthodoxpredictionofdivergingU.S.andFrenchemploymentperformancefaresevenworseifabettermeasureofyouthunemploymentisused.Theconventionalmeasureisdefinedastheunemployedshareofthelaborforce(theunemployedplustheemployed).ButunlikeU.S.students,moststudentsinFrancedonotwork,andthisisnotsimplyareflectionoftheabsenceofjobopportunities.Thesamewastrueinthe1960s,whentheFrencheconomywasatnear-fullemployment(HowellandOkatenko,2010).Withloweremployment,thesamenumberofunemployedwilltranslateintoahigher
U/LFUS
U/POPUS
U/LFFR
U/POPFR
5
10
15
20
25
30
1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013
%
U/LFUS
U/POPUS
U/LFAUS
U/POPAUS
6
8
10
12
14
16
18
20
1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013
%
33
unemploymentrate,byconstruction.Amuchbettermeasureofunemploymentforyoungpeopleistheunemployment-to-populationrate,whichisshownforFrenchandAmerican15to24yearoldsatthebottomofFigure7.TheseFrenchandU.S.unemploymentrateshavetrackedeachothercloselysincethe1980s,bothfluctuatingbetween6percentto10percent.ThereisclearlynoevidenceofthepredicteddivergenceinFrench-U.S.employmentperformance.Figure7bpresentsthesameunemploymentdatafor15-24yearoldsintheUnitedStatesandcomparesthemtoAustralia.LikeFrance,Figures5,6aand6bshowhigh,andinthecaseofFigures6aand6b,stronglyrisinginflationadjustedvaluesoftheAustralianminimumwage.ThisshouldhaveledtohigheranddivergingunemploymentratesforAustraliarelativetotheUnitedStates.ButFigure7boffersnosupportforthisconventionalprediction:TheAustralianconventionalunemploymentrate(U/LF)fellsharplybetweentheearly1990sandtheglobal2008economiccrisis,tolevelsbelowtheUnitedStates.Asimilarpatterncanbeseenusingunemployment-to-populationrates.Themostrecentdatashownearlyidenticalyouthunemploymentratesonbothmetrics(despitetheeffectsofacollapseincommoditypricesontheAustralianeconomysince2012).Figure8:TheIncidenceofLowPayand2015EmploymentRatesforYoung(25-34)Less-EducatedWorkersfor17Countries
Sources:OECD“LowPayIncidence”dataextractedfromOECD.Stat,June1,2016,andarefor2013excepttheNetherlands(2006),France(2005),andNorway(2009),whicharefromSchmitt (2012);theloweducationemploymentrateisfor25-34yearoldswithlessthanuppersecondaryschooling,(EO-OECD2015,tableE).
USUK
CA
IRE
GER
NL
AUT
SP
AUS
POR
DEN
GRE
NZ
FRFNL
NOR
ITBEL
40.00
45.00
50.00
55.00
60.00
65.00
70.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00
LowEdu
EmployRate
LowWageShares
34
AnotherperspectiveonFrench,AustralianandU.S.employmentperformanceisprovidedbyFigure8,whichshowsascatterplotofthelow-wageshareofemploymentagainstthelow-educationemploymentrate.Ahigherminimumwage,togetherwithhigherratesofcollectivebargaining(amongotherfactors)explaincross-countrydifferencesintheincidenceoflowpay.Ifthese“labormarketrigidities”priceworkersoutofthelabormarket,thenreducingthelow-wageshareshouldalsoreducetheloweducationemploymentrate,asyounglesseducatedworkershaveahardertimefindingandkeepingjobs.ButFigure8showsnocross-countryrelationshipbetweentheincidenceoflowpayandthelow-educationemploymentrate.Indeed,whilethereisa14percentagepointgapinthelow-wageshareofemploymentbetweenFrance(11percent)andtheUnitedStates(25percent),theemploymentratesforyounglesseducatedworkersisnearlythesame.Similarly,Australia’sincidenceoflowpayis9percentagepointsbelowtheU.S.level,butwithlow-educationemploymentratesabout7pointshigher.WehavealsohighlightedDenmark,whichshowsthestrongestchallengetotheorthodoxprediction—alow-wageshareofemploymentofjust8percent,17pointsbelowthe25percentrateoftheUnitedStates,butDenmarkstillshowsasuperiorlow-educationemploymentrateforyoungworkers.RelyingonabroadermeasureofhowtheFrenchandU.S.economiesperformforyoungworkers,Figure9comparestheNEETrate(NotinEmployment,EducationorTraining)for20-to-24yearolds.Ifyoungpeoplenotattendingschoolare“pricedout”ofthelabormarketbyahighminimumwagethenweshouldexpectaveryhighandrisingNEETrateforFrancecomparedtotheUnitedStates.Figure8showsthatFrancehasonlyslightlyhigherNEETratesthantheUnitedStates,andthegaphasclosedsignificantlysince2000,from3.2pointsin2000(17.6percentforFranceand14.4percentfortheUnitedStates)tojust0.6points(19.4percentcomparedto18.8percent).WhiletheFrenchNEETrateincreasedby1.8pointsoverthese13years,theU.S.rateincreasedby4.4points.Again,basedonthelevelsandchangeintherelativevalueoftheminimumwage,fromaconventionaltextbookperspectiveweshouldbeseeingdivergence,notconvergence.Insum,thiscross-countryevidenceoffersnosupportfortheconventionalviewthatahighrelativevalueofthelegalwagefloor(theKaitzindex)offersagoodguidetotheabilityofaneconomyto“afford”amuchhigherwagefloor(whethersetbuya
35
statutoryminimumwageorbycollectivebargaining).IftheKaitzratioturnsouttobeagoodbenchmark,itwouldhavetobeintheregionwellabove60percent.39Figure9:TheShareofFrenchandAmerican20-24YearOldsNotEmployed,EnrolledinSchool,orinTraining(the“NEETRate”),2000and2013
Source:OECD(2015),Table3.4.c) TheSector-LevelKaitzIndex:ABetterNJLMetric?
WhilethemedianwageforageographicareaseemsapoorguidetosettinganNJLwagefloor,abetterjobmightbedonebyusingthemedianwageforanarrowlydefinedlow-wagesector(suchasretailtradeorfoodservices),becausetheconditionsfacingemployersthatmatterforthewage-employmentrelationshipmaybefairlysimilar.TheexperienceoftheUnitedKingdomspeaksdirectlytothequestionofhowhightheKaitzratiocanbeinthelow-wagesectorswheremostminimumwageworkersareemployed.Table1showsthatthecountry’sNationalMinimumWage(NMW)-to-medianratioinalllow-payingsectorsincreasedfrom67.5percentin1999to80.2percentin2015(secondtolastrow).Forspecificsectors,overthese16yearstheNMW-to-medianratiorosefrom81.9percentto92.5percentincleaning,from78.6percentto88.1percentinhospitality,andfrom69.5percentto79.5percentinretail.This
39ItshouldbenotedthatwearereferringtotheOECD’sKaitzratios,whicharemeasuredbythefull-timemedian, amore stringent standard than the targetsproposedby theUKgovernment and theUK’sResolutionFoundationforawagefloorthatis60percentoftheoverallmedian.
17.6
14.4
19.4 18.8
0
5
10
15
20
25
NEET. France NEET.U.S.
2000
2013
+1.8 +4.4
36
verysubstantialwagecompressionoccurredwithoutevidenceofnegativeemploymenteffects(LowPayCommission,2014,p.12).ThisevidenceindicatesthattheNJLKaitzratioforlow-payingsectorsintheUnitedKingdomisabove80percent,the2015value.Howmuchabove?HadtheLowWageCommissionadoptedthebackward-looking“unchartedwaters”NJLcriterion,evidencewouldhavebeenrequiredtoconfirmthateachoftheseincreasessince1999wouldnottriggerjobloss.Intheabsenceofsuchevidence,wewouldnotknowthattheseincreasesintheNMW-to-medianratioof10-to-15percentagepointscouldtakeplacewithoutanyjobloss,muchlesslarge-scalejobloss(aswaspredictedinthe1990s)—andmanyU.K.workerswouldbemuchworseoffasaconsequence.40Table1:TheUK’sKaitzRatioforLow-WageSectors,1999-2013Sector 1999 2008 2012 2013 2014 2015 Highestbite
Cleaning 81.9 90.1 93.5 92.5 92.7 92.5 93.5 2012Hospitality 78.6 85.3 86.9 88.1 87.7 87.7 88.1 2013Hairdressing 83.5 80.4 85.8 84.4 85.1 86.7 86.7 2015Childcare - 69.6 82.8 84.2 83.7 84.8 84.8 2015Retail 69.5 76.7 79.5 78.1 79.4 79.1 79.5 2012Socialcare 60.8 72.2 76.8 78.4 78.7 79.5 79.5 2015Agriculture 67.5 71.7 75.1 71.8 72.1 73.2 75.1 2012Textiles 62.1 69.9 71.7 71.0 71.0 73.0 73.0 2015Leisure 59.3 66.8 69.5 70.8 71.1 72.1 72.1 2015Employmentagencies
- 67.7 68.0 68.1 71.5 70.4 71.5 2014
Foodprocessing 55.6 65.2 70.4 68.4 70.0 72.2 72.2 2015
Low-payingsectors 67.5 75.5 79.4 78.9 79.6 80.2 80.2 2015
Nonlow-payingsectors
42.2 45.6 46.0 45.9 46.2 47.0 47.0 2015
All 47.1 51.2 52.8 52.5 53.2 54.1 54.1 2015
40AccordingtotheUK’sLowPayCommission(2014),itispossiblethattheremaystillberoomforhigherincreasesinthelow-paysectorKaitzratioswithoutriskofjoblossifthenationalminimumwageisincreasedgradually.“Thebite(Kaitzratio)isatornearitshighesteverlevelinthesesectors.AgainstthatbackgroundourviewisthatintheseconditionsanincreaseintheNMWsuchastocausealargeriseinthebitewouldrunahighriskofadverseemploymenteffects.Thatriskwouldbemoreacuteifanincreaseweretotakeplaceveryquickly:theevidencefrompastincreasesintheNMWisthattimetoadjustbusinesspracticeinordertomanageadditionalwagecostshasbeenimportantinenablingemployerstocope.Wecontinuetoreceiveadvicefromemployersthataverybigrisethattakesplaceinonegowouldbethehardestforthemtoabsorb”(LowPayCommission2014,pp.31-2).
37
Source:LowPayCommission(2016).TheResolutionFoundationestimatesthattheincreaseinthebitefromtheNMWin2014totheNationalLivingWage(NLW)in2020wouldbefrom83percentto98percentintheretailtradesector,from93percentto110percentinaccommodationsandfoodservices,andfrom85percentto101percentincleaningservices.41Theincreaseinthewagebillofthesethreesectorsisestimatedtobe2percent,3.4percent,and3percentrespectively.Evenintheselabor-intensivesectors(becausethewagebillisonlyapartoftotaloperatingcosts)onlyafractionofthis2-to-3.4percentwageincreaserangeneedstobeaccommodated,andasboththeLowPayCommissionandtheResolutionFoundationhaveindicated,U.K.employershaveshownthattheycanrespondnotnecessarilyoronlybycuttinghoursandjobs,butalsoviapriceincreases,reducedturnovercosts,higherproductivity,andlowerprofits(D’ArcyandCorlett2015,Table1).Insum,theexpectedchangeinthewagebillafterlikelyadjustmentstocopewithamandatoryincreaseinthewagefloorisamuchbetterguidetoexpectedjoblossthantherelativemedianwage.ButiftheKaitzindexistobeusedastheguide,itisprobablybestcalculatedatthesectorlevelinappropriatelocalities.TheU.K.evidencesuggeststhatveryhighsector-levelKaitzratiosareconsistentwithlittleornojobloss.6.TheFederalWageFloor:AMinimumLivingWage
Everyworkershouldbeensuredaminimumwagewhichwillenablehimorhertomaintainabecomingstandardoflifeforhimselfandhisfamily.ApartaltogetherfromconsiderationsofhumanityitisonthehighestinteresttotheStatethatchildrenshouldbebroughtupunderconditionsthatwillmakethemfitandefficientcitizens.BritishPrimeMinisterLloydGeorge,191942
41TheestimatedratiosoftheNLWtothesectormediancanbeover100percentbecausetheFoundationdidnotestimatethefuturemedian;theobjectivewasonlytoshowthechangefromtheNMWbitein2014towhatitwouldbein2016and2020undertheNLWwithoutchangingthemedian.ButclearlytheNLWwouldcontinuetheconvergenceoftheoftheKaitzratiotowards1inverylow-wagesectors.42QuotedbyWaldman(TheIncidenceofLowPayand2015EmploymentRatesforYoung(25-34)Less-EducatedWorkersfor17Countries2004,p.196).
38
Theconcernoverjoblosshasalwaysplayedacentralroleintheminimumwagedebate,butuntilrecentlytheunderlyingmotivationwasthatexpressedin1919byLloydGeorge(above)andintheopeningsentencesofthe1938FairLaborStandardsActasnotedabove.Andintheolderrhetoric,patriciannationalpoliticalleaderscouldspeakofthe“nationalevil”ofthepaymentofpoverty-levelwages(Churchill)andthatfirmsthatdidsohad“norighttocontinueinthiscountry”(Roosevelt).43Itisnotablethatthisearlierframingtookplaceinaperiodwithvirtuallynosocialsafetynetandnojobprotectioninextremelycompetitivelabormarkets.AnymandatedwagefloorcouldbeexpectedtoresultinjoblossinU.S.andU.KeconomiesthatwerestrugglingwithasurpluspooloflaborthatthatproducedwhattheU.S.economistJohnBatesClarkcalled“hungerdiscipline.”Thisearly20thcenturydebatewasframedintermsofalivingwage.Bycontrast,intoday’sdebate,inacontextofmuchlowerunemploymentandsubstantiallygreatersocialprotection(howeverinadequate),thediscourseisdominatedbywhatthestatisticalevidencesaysabouttheeffectsofincreasesinthestatutorywageflooronjobloss.Withthebestevidencenowshowingnodiscernibleemploymenteffects,manyofthestrongestadvocatesforsubstantialhikesintheU.S.federalminimumwagehavemadethecaseonNoJobLoss(NJL)grounds.Indeed,theargumentisthatnotonlywillworkersnotlosejobs,buttherewillbelittleornoharmdonetoanyone—aperfectexampleoftheeconomist’s(near)Pareto-improvement:manygainalotandnooneloses.Settingasidethemeritsofthisview,whatisstrikingaboutthisframingisthattheultimateprogressivegoal(alivingwage)andthestrongRooseveltianrhetoricinsupportofithasallbutdisappeared.Thisleavesthecaseforminimumwagethatcanmeaningfullyimprovethelivingstandardsofworkingfamiliessubjecttothevagariesofthestatisticalanalysisovertheriskthatsomepoverty-wagehigh-turnoverjobswilldisappear,withnoplaceinthepolicydebatefortheethicalandefficiencypayoffsthatusedtobefrontandcenter.Inthissection,wesuggestthatethicalandefficiencyconsiderationsshouldbereintroducedtothedebate.Theprogressivecaseforasubstantialincreaseintheminimumwageshouldbereorientedfroma“no-harm”(NJL)framingtoabenefit-basedonethatexplicitlycallsforaMinimumLivingWageonbroadlydefinednet-benefitgrounds,whichincludenotjustthenetmonetarybenefitsofahigherwageforthestandardoflivingofworkingfamilies,butalsothemanypositivespillovereffectsofa“high-road”employmentmodel.Decentpayhelpsworkingfamiliesavoiddependenceonpublicspendingthatisstigmatizingandpoliticallydivisive,and
43Seethetitlepageforthefullquotes.
39
wouldhelpendthecurrentpracticeofsubsidizinglow-wage,“race-to-the-bottom”employmentmodelsthathaveincreasinglycharacterizedthehumanresourcepracticesoffor-profit,non-profit,andgovernmentemployersalike.AlthoughwedonotattempttoidentifyaparticularMLWlevelorthedateatwhichitshouldbefullyphasedin,weoffersomeevidenceofthekindthatmightbeusedtodoso.a)TheProblem:ALow-WageSocialModelTheresponsibilitythatathestatutorywagefloormustbeartobringtheincomeofworkingfamiliestolevelssufficienttoprovideadecentstandardoflivingdependsonthelargerinstitutionalcontext.Thisincludesthenatureofsocialprotectionpolicy(e.g.,targetedtaxbenefitsliketheEarnedIncomeTaxCredit,andchild,housing,education,andhealthbenefits),theextentandeffectivenessofcollectivebargaining,andthesecurityandbargainingpowerthatcomeswiththestrengthoflabormarketinstitutionssuchasemploymentprotectionlawsandunemploymentbenefits(Grimshawetal.2016;MarchalandMarx2015).Inaddition,corporategovernanceandwage-settingnormsintheprivatesectorcanplayabigroleinthesettingofwagesforless-skilledworkers.Onalloftheseinstitutionaldimensions,U.S.workersfacetheskimpiestsocialsafetynetandthemostcompetitiveandprecariousjobmarket.Todate,thefederalminimumwagehasbeensettoalignwiththis“low-road”labormarketmodel,witharealandrelativewagefloorthatisthelowestintheaffluentworld(seeSection2).ThefailureoftheU.S.labormarket—andthefederalminimumwage—tosetadecentlowerboundaryforhourlypaycanbeseeninthedata,whichshowssteadilyworseningperformancesince1979,particularlyforyoungworkers.Defining“lousyjobs”asthoseinwhichworkerswerepaidlessthan2/3ofthemedianwageforfull-timeworkersages18to64($12.50in2014)orwereworkinginvoluntarilypart-time,theincidenceoflousyjobsforyoungworkersages18to34withoutacollegedegreehasincreasedastronomicallysince1979.Forfemales,thelousy-jobrateforthisgroupincreasedfrom53.1percentin1979to70.1percentin2014;forsimilaryoungmen,theincreasewasevenlarger:from28percentin1979to57.1percentin2014.Thiscomparestolousy-jobratesforyoungmenwithoutacollegedegreeof44.9percentintheUnitedKingdomandjust18.1percentforFrance(2012).ComparedtotheU.S.youngfemalerateof70.1percent,theyoungU.K.andFrenchfemaleratewere59.9percentand29.2percent.4444Author’scalculations(Howell’sEquitableGrowthDecentJobsProject).Canada’slousy-jobratefortheseworkerswasalmostasbad:68.6percentforfemalesand47.5percentformales.
40
b)From‘NoJobLoss’toa‘TotalEarnings’andMinimumLivingWageStandard
Wesuggestthattheprimaryconsiderationforsettingthefederalwageflooristhestandardoflivingthatcanbeattainedfrompaidemployment.Wedonotproposeaspecificdetailedplanforsettingthefederalwagefloor,muchlessaparticularMinimumLivingWage(MLW).ButwedoarguethattheprimaryconsiderationinthesettingoftheMLWisasociallyacceptabletargethourlywageonstandardoflivinggrounds,andthatemploymenteffectsoughttobeanimportantbutsecondaryconsideration.Inasking“Cantheminimumwagebe‘toohigh?”,thedominantminimumwageresearchersofthelastgeneration,DavidCardandAlanKrueger,implyaconventionalNo-Job-Loss(NJL)criterion.
Ultimately,however,aminimumwagethatissettoohighwouldbeexpectedtocauseemploymentdeclines,evenwhenfirmshavemarketpowerandsetwagesmonopsonistically.Ourviewisthatthepoliticalprocessusuallypreventstheminimumwagefromexceedingthepointwhereitadverselyaffectstotalemployment,butitisimportantforresearchtoestablishwheresucheffectswouldoccur…Eveniftheminimumwagedoesexceedthislevel,however,itwillstillincreasetotalearningsforlow-wageworkersiftheelasticityofdemandislessthanoneinabsolutevalue(CardandKrueger2015,p.xx).
“Toohigh”isdefinedinthispassageasalevelthatcausesemploymentdeclines—alevelabovewhatwehavecalledtheNJLthreshold.Buttheyalsoappeartoenvisionanotheryardstick,anarrowlydefinedmeasureofnetbenefits:thechangeintotalearningsforlow-wageworkers.ThiswouldgofarbeyondthestrictNJLcriterion,inwhichthewagefloorshouldbesettoprecludetheriskofanyjoblosstakingplaceanywhere.Sinceitiswell-knownthattheelasticityoflabordemandisfarbelowone—Manning(2016)hasarguedthatitmaybeessentiallyzeroevenforU.S.teenagers—therewouldbeagreatdealofroomunderthis‘totalearnings’standardforwagefloorstobesetwellabovetheNJLthreshold.ThecaseforthismoreambitiouswagefloortargetwouldbeevenstrongerifweheededtherecommendationsmadebyJohnBatesClarkin2013andcloselymonitoredemploymenteffectsinnear-real-time,with“emergencyrelief”legislatedalongwiththeminimumwageincreasestosupportworkerswhowereharmed.
Emergencyreliefneedstoaccompanytheminimum-wagelaw,andeffectivemeasuresforitmustbereadytoactthemomentthelawis
41
passed.Itwillnotdotodischargetheworkersandthendebatethequestionastohowbesttogivethemwork.Moreover,suchemploymentaswefurnishshouldbesuchasself-respectingpersonsmayproperlyaccept(quotedbyPrasch,2000,pp.257-8).
Adoptingthe“totalearnings”criterion,theMLWshouldbesetwiththegoalofeliminatingextremelowpay,asBelgium,France,andtheScandinaviancountries(throughcollectivebargaining)havealreadyaccomplished,andifthisshouldleadtosomejobloss,thereisnoquestionthatthenetbenefits(evenifmeasuredonlyintermsofearningseffects)wouldvastlyoutweighthecostsofgenerous“emergencyrelief.”InsteadofinvokingthenarrowParetoCriterionofnoharmtoanyone,thiswouldmeantheadoptionofwhateconomistsrefertoastheCompensationCriterion,inwhichnetbenefitscanbeusedtofullycompensatethoseharmedbyapolicy.TheLowPayCommissionintheUnitedKingdomoffersamodelofaninstitutionalsetupthatcouldoperationalizeanMLW,onethatisinextricablylinkedtoacompensationschemethateffectivelyensuresfullemployment.45Thiscommissionwouldberesponsiblefor(1)decidingtheappropriateMLW;(2)regularmonitoringandadjustmentsofthesemandatesbasedonevidenceoftheeffectsonbothlivingstandardsandemployment;and(3)planningeffectiveresponsestoanyjoblossesthatcannotbeavoided.46c)TheMLW:HowLittleisTooLittle?AlthoughonlyahandfulofcountrieshadestablishedminimumwagesatthetimeoftheTreatyofVersailles(1919),thetreatycalledfor“thepaymenttotheemployedofawageadequatetomaintainareasonablestandardoflifeasthisisunderstoodintheirtimeandcountry”(Anker2011,p.16).Similarly,theopeningsentencesoftheFairLaborStandardsActof1938makeclearthattheprincipalmotivationforlegislatingaminimumwagewastoeradicatethose“laborconditionsdetrimentaltothemaintenanceoftheminimumstandardoflivingnecessaryforhealth,efficiency,andgeneralwell-beingofworkers.”TheminimumwagewasalsoincludedinthegeneralprinciplesoftheInternationalLabourOrganization(ILO);aresolutionadoptedin1945recommended“theestablishmentofappropriateminimumwagestandards,adequateforsatisfyingreasonablehumanneeds”inorderto“assisttheprogressiveraisingofthestandardoflivingofallworkers”(ILO2014,p.3).45AstheResolutionFoundation(2014)haspointedout,theLowPayCommissionshouldhavebeencalledthe“MinimumWageCommission”becauseit’schargewasnottoreducetheincidenceoflowpay.46AstheAmericaneconomistJohnBatesClarkwroteoveracenturyagoin1913(Prasch2000).
42
Thereisnoconsensus—nationallyorinternationally—abouthowtodefineaspecificandrealisticlivingwage.Mostapproachesagreethatataminimum,thelivingwageshouldbehighenoughtoallowaworkerwithafamilytocoverthebasicsubsistencecostsofshelter,food,transportation,clothing,andutilities.Itemssuchashealthcareandeducationarefreeinsomecountriesandnotinothers,sothatimpactsthenecessarywage.TheILO’sRichardAnker(2011)hasdevelopedamethodologyfordeterminingalivingwageinaninternationalcontext.PerhapsthelivingwagecalculationsfromabroadthataremostdirectlyrelevantforthinkingaboutanappropriateMLWfortheUnitedStatescomesfromtheLivingWageFoundationintheUnitedKingdom,whichreliesonresearchfromtheCentreforResearchinSocialPolicy(CRISP)atLoughboroughUniversity.Theirliving-wageestimatesareintendedto“showhowmuchhouseholdsneedinaweeklybudgetandhowmuchtheyneedtoearninordertoachievethisdisposableincome”foravarietyofhouseholdtypes.47TheFoundation putsthe2016livingwageat£8.25outsideLondonand£9.4inLondon.48Thesefigureswouldtranslateintoa2016U.S.livingwageofbetween$11.77and$13.41.49IntheUnitedStates,thereareseveralmethodologiescommonlyusedtoestimateabasic-needsbudgetforworkerswithdifferentfamilytypes(e.g.singleadult,singleadultwithonechild,twoadultswithtwochildren).50Mostfindthatthewageneededtopaythebasiccostsofliving—housing,food,transportation,utilities,taxes,healthcare,savings,clothing,andpersonalitems—requiresafull-timejobatawagethatis,asFigure10suggests,atnearly$14perhourforasingleperson,andatleast$22forasingleadultwithonechildinlowcost-of-livingareas.Forexample,amongtheninecitiesshowninthefigure,Minneapolishasthelowestbasic-needsbudgetforasingleperson($13.62)andHoustonhasthelowestforasingleadultwithonedependentchild($22.67).Thiscomparestotheproposed2020federalminimum
47http://www.lboro.ac.uk/research/crsp/mis/whatismis/48FortheLivingWageFoundation,seehttp://www.livingwage.org.uk/;ForCRISP,seehttp://www.lboro.ac.uk/research/crsp/49Theseareroughestimates,calculatedbyusingtheOECD’sPurchasingPowerParity(PPP)multiplier(1.426)for2014totheLivingWageFoundation’sestimateoftheLondonandoutsideLondon2016livingwage.The1.426factorwascalculatedbytakingtheratiooftheU.K.PPPadjusted2014nationalminimumwagefromtheOECD($9)totheUnitedKingdom’s2014nationalminimumwageinpounds.50ThisincludestheEconomicPolicyInstitute’sFamilyBudgetCalculator,theSelf-SufficiencyStandarddevelopedbyDianaPearce,andtheMITLivingWageCalculatordevelopedbyAmyGlasmeier.Alloftheseprovideestimatesoftheincomeneededtocoverbasiclivingcosts,byfamilysizeandtypeaswellascityandstate.AlsoseeFredericksen(2015).
43
wageof$12,whichistheequivalentof$10.92in2016(basedonCBOinflationprojections),andtheproposed$15in2021—about$13.34intoday’sdollars.Figure10:TheFull-TimeHourlyWageRequiredforBasic-NeedsBudgetbyFamilyTypeforSelectedCitiesfor2016(withlinesshowingproposed2020and2021minimumwagesin2016dollars)
Source:Tungetal.(2015),table3.1;derivedfromEPI’sFamilyBudgetCalculator).Figure11:TheFull-TimeHourlyWageNecessarytoRentaModest2-BedroomApartment:tenselectedstates,statewideaverageandthenonmetropolitanaveragein2015(withlinesshowingproposed2020and2021minimumwagesin2016dollars)
Source:NationalLowIncomeHousingCoalition(NLIHC2015).NLIHCcalculatestheseestimatesbasedonHUD’spublishedFairMarketRentandassumesthathousingcostsare30percentofincome.AsecondsourceforbenchmarkingtheMinimumLivingWageistheevidencefromtheNationalLowIncomeHousingCoalition.Torentamodest,two-bedroomapartment,theaveragewagerequiredforafull-timeworker(40hoursand52
14.64 14.10 13.45 13.91 13.6215.67
14.0317.00
21.0723.59 24.06 24.90
22.67
26.76 26.40 25.86
29.58
39.35
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
Bakersfield Phoenix Colorado<Spr Houston Minneapolis Chicago Buffalo Baltimore Washington
Single<Adult With<1<Child
2021: $15<($13.34)2020:<$12<($10.92)
0.002.004.006.008.00
10.0012.0014.0016.0018.0020.0022.0024.0026.0028.00
Alabama Arizona California Minnesota Montana New:York Ohio Oregon Pennsylvania Tennessee
State Combined:Nonmetro:Areas
2021: $15:($13.34)2020::$12:($10.92)
44
weeks)was$19.35perhour,or$15.50foraone-bedroomunit(assumingrentsare30percentofhouseholdincome).
AsFigure11shows,evenforruralAlabama,therentrequiredforaminimallydecenttwo-bedroomapartmentinnon-metropolitanAlabamaisestimatedtorequirea$12.18hourlywageforafull-timeworker,whichisabout$1.20morethantheproposed$12wagefor2020wouldprovide($10.90in2016).TheaverageruralAlabamaworkercouldaffordthisapartmentwithawageequivalenttotoday’svalueoftheproposed2021$15wage,butwithonly$1leftoverperhour($13.34vs$12.18).However,eventhe$15proposal($13.34today)wouldbetoolittletocovertherentalofamodesttwo-bedroomapartmentinthefivemostexpensivecitiesinAlabama,whichwouldrequireafull-timehourlywagerangingfrom$15.15to$15.63(NLIHC2015,p.16).Thesecost-of-livingestimatessuggestthata2020wagefloorof$12wouldnotcomeclosetosatisfyingareasonableMLWstandard—evenforruralAlabama.Indeed,theysuggestthattheMLWwouldneedtobeatleast$14in2016dollars,andperhapsintherangeof$16.00to$17.00by2021.
Additionalevidenceonbasic-needsbudgetscanbefoundinthe“MakingWorkPay”reportsbytheNationalCenterforChildreninPoverty(NCCP),whichprovideestimatesoftheincomefamiliesrequireforbasicneedsincitiesofselectedstates.Thisisaparticularlyvaluablesourceforwhatisneededfromawageaftertakingintoaccounttheavailabilityofcity,county,state,andfederalmeans-testedsocialsupport.Forexample,themostrecentfindingsforMontana(for2010)suggest:
AcrossthesevenlocalitiesexaminedinMontana,familiesneedincomesofovertwicethefederalpovertyleveltocovertheirbasicexpensesofhousing,food,transportation,healthinsurance,childcare,andothernecessities….(thefigure)showsthatasingleparentwithtwochildren,onepreschool-agedandoneschool-aged,needsanannualincomerangingfrom$39,000inHavreto$46,000inKalispelltocovertheseexpenses.Thisisequivalenttoawageof$19to$22perhour—twotothreetimesthevalueoftheminimumhourlywageof$7.35andoneandahalftimesthevalueofMontana’smedianwageof$13.65(Chau2011,p.4).
TheNCCP’sstudyofIowa(for2008)producedsimilarresults.InbothIowaandMontana,thecombinationoflocal,stateandfederal“work-support”programscanpotentiallymakeupformost,orevenall,ofthegapbetweenthesebasicexpensesandnetearningsifthe“take-uprate”is100percentinareaswiththemostcomprehensivesetofsupportiveservices.Buteligibilityforthesepublicsubsidiesvariesovertimeandbyjurisdiction,andrequiressubstantialtimeandenergytoknowwhatisavailable,toshoweligibility,andtoapplyandcollect.Asaresult,
45
accordingtoNCCP,“manyfamiliesdonotaccessalloftheprogramsforwhichtheyareeligible.”51
d) NetImpactsforWorkersIftheminimumwageisraised,workersmayloseeligibilityforsomeprograms,suchastheEarnedIncomeTaxCredit(EITC),theSupplementalNutritionAssistanceProgram(SNAP),orfoodstamps,52childcareandhousingassistance,andmedicalcare.Atthesametime,policycanbeusedtooffsethigherlaborcostsforemployers(suchastaxpolicyorsubsidies).Inordertodeterminethe“right”federalminimumwage,itiscriticaltotakeintoaccountthenetoutcomesforbothworkersandemployers.Thispointishighlightedwhencomparingminimumwagesintheinternationalarena.AccordingtoOECDestimates,theaveragenetU.S.minimumwagewasjust$6.26in2013,about$1.00lessthanthegross$7.25wage.53Thiscomparestoanetwageof$7.06fortheUnitedKingdom,$7.18forCanada,$8.24forFrance,and$9.54forAustralia(OECD2015a).TheOECDestimatesthatthenetgain—theshareoftheminimumwageincreasethataworkertakeshomeaftertaxesandbenefitreductions—toaU.S.minimumwageworkerwasbarelyover40percent(40centsforeachdollarincreaseintheminimumwage)in2013,whichwasmorethaninAustralia(32percent)butslightlylessthanFrance(45percent)andfarlessthanCanada(over60percent)andtheNetherlands(over80percent)(OECD2015b).GoodillustrationsofthepotentialfortheseoffsettingeffectscanbefoundintheNCCP’s“MakingWorkPay”reportsforIowaandMontana(discussedabove).Bothshowasubstantial“cliffeffect”whereworkingfamilyincomesactuallyfallastheworker’shourlypayincreases.InthecaseofIowa,asFigure12shows,incomewoulddropforafamilyofthree(asingleparentandtwochildren,ages2and6)inDesMoinesbynearly$8,000duetothelossoffoodstampsandachildcaresubsidy
51AccordingtotheChau’sNCCPMontanareport(2011,p.6),“ArecentpaperfromtheUrbanInstituteshowsthatnationallyoneinfiveeligiblechildrendonotparticipateinMedicaidorChildren’sHealthInsurancePrograms(CHIP),nearlyfourin10eligibleworkinghouseholdswithchildrendonotreceiveSNAP,andsevenin10familieseligibleforchildcaresubsidiesarenotserved.5Thesefindingsechothoseofapreviousstudy,whichfoundthatonlyfivepercentoflow-incomeworkingfamiliesreceivedabroadworksupportpackageofpublichealthinsurance,SNAPandachildcaresubsidy.”52SupplementalNutritionalAssistanceProgram.53TheOECDestimatesthetaxesthatworkersmustpay,includingmandatorycontributionstoretirementpensions,incometaxes,unemploymentinsurance,andothersocialprograms.Thenetminimumwageistheestimatedtake-homepayforafull-timeminimumwageworker.WagesareadjustedusingPurchasingPowerParitiesforprivateconsumption(OECD2015a).
46
astheworker’spayincreasedfrom$10.50to$12(2008)(Fassetal.2008,figure1).AsimilarfamilyinBillings,Montanawouldhaveseenadropofabout$7,000in2010fromawageincreasefrom$13to$13.50duetothelossofthechildcaresubsidy.Inaddition,anincreaseinpayfrom$15to$18wouldincreaseannualincomebyjust$1,000becauseofthelossoffoodstamps)andotherbenefits(Chau2011,figure4).Theseexampleshighlighttheneedforacarefulassessmentofthecorrectminimumwagebasedonlocality,andavailablepolicies.Italsoshowshowitiscrucialthattheminimumwagebeunderstoodaspartofapackageofpoliciesinadditiontothewage.TheNCCPreportprofilespotentialimpactsforhypotheticalfamilies,butwealsohavesomeresearchontheactualimpactofincreasedwagesonbenefiteligibilityandnetoverallearningsfromresearchonmunicipallivingwageordinances.Researchersconductedinterviewswithworkerscoveredbythelivingwageordinancetocomparetheirbefore-and-afterincome,includingrelianceonsocialprograms.Figure12:NetFamilyResourcesasEarningsIncrease:DesMoines,IA(SingleParentwithtwoChildren,Ages2and6)
Source:NationalCenterforChildreninPoverty(2008).Forexample,Fairrisetal.(2005)studiedtheimpactoftheLosAngeleslivingwageordinance,whichrequiredfirmscoveredbytheordinancetoraisewagesfromaminimumof$4.25to$7.25perhour(plushealthinsurance)in1996.54TheyfoundthatafterthelivingwagewasimplementedinLosAngeles,amajorityofworkers
54Thelivingwageisadjustedeachyearwiththecostofliving.By2015theL.A.livingwageratewas$11.17plushealthbenefits(http://www.lawa.org/welcome_LAWA.aspx?id=596).
47
reportedlessincomefromsocialprogramsbutanetincreaseinannualincome(Fairrisetal.2005).Thegainsweregreatestforsingleormarriedworkerswithoutchildren,astheywereeligibleforfewerprogramstobeginwith.Singleparentswithchildrenwere,onaverage,themostdependentongovernmentprograms,andthereforeexperiencedthebiggestdropineligibility.Still,workersinterviewedreportedanetgaininincome.Thenetgainwasnotenoughtoliftmanyworkersoutofpoverty,giventhatthelivingwageratewassetjustatthefederalpovertyline,whichsuggeststheneedforahigherwagelevel.Toensurethatminimum-wageworkersgetthemostofanyincrease,appropriateadjustmentstothebenefitschedulesofprogramssuchastheEITCandSNAPshouldbecentraltothedebateovertheappropriatelevelofthefederalminimumwage.557.ConclusionTheUnitedStatesisjustemergingfromaglobalfinancialcrisisthathaddevastatingconsequencesformanyU.S.workers,mostvisiblyinjoblessnessandthecollapseofhousingvalues.Somesevenyearsafterthetrough,theemploymentrateisonlyslowlyrecovering,buttheconventionalunemploymentrateisnowaround5percent,halfwhatitwasin2009-10.Astheemploymentcrisisrecedes,whatremainsisthemuchlongertermcrisisinlowpay,onethatshowsasteadyworseningsincethelate1970s.AccordingtotheBureauofLaborStatistics,therearemorethan35millionpeoplewhoare“workingpoor.”Theshareofworkingpoorwentfrom5.1percentofthelaborforcein2007to7.2percentin2010,andhaschangedlittlesincethen,despitethetepideconomicrecovery.Familieswithchildrenunder18yearsoldwerethreetimesmorelikelytobeamongtheworkingpoor.Theseratesofworkingpovertyarethehighesttheyhavebeensincethe1980s,whentheBLSstartingcollectingdata(BLS2015).56Theproblemofin-workpovertywillnotbesolvedbysmall-scaletinkeringwithourcurrentlabormarketpoliciesandinstitutions.Workersrequireamassiveinfusionofbargainingpower.IntheabsenceofeffectivecollectivebargainingonanorthernEuropeanscale,theinterventionthatcanbemosteffectiveintheneartermisasubstantialhikeinthenationalwagefloor,onethatrequiresemployers—for-profit,non-profitandgovernmentalike—topayadecentlivingwage.
55Onhowtobestcombinethesepolicies,see,forexample,Wicks-LimandPollin(2012).56Theworkingpooraredefinedasdefinedaspeoplewhospentatleast27weeksinthelaborforcebutwhosehouseholdincomesstillfellbelowthefederalpovertylevel.Giventheinadequacyofthefederalpovertylinemeasures,theactualrateofworkingpoorislikelymuchhigher.
48
Whilethelivingwagemovementhasmadegiantstridesincitiesandstatesacrossthecountry,thefederalminimumwageremainsatjust$7.25,alevelfarbelowthatofmostotheraffluentcountrieswithstatutorywagefloors,inbothreal(buyingpower)andrelative(tothemedianwage)terms.Yet,evenamongmanyadvocatesforahigherfederalminimumwage,thegoalsseembarelyadequate,withtargetsof$10.10,$12,oreven$15thatarenottobefullyphasedinuntil2020-23,dependingontheproposal.Afterall,AustraliaandFranceeffectivelyoutlawedlowpay,definedbywagesthatarelessthantwo-thirdsofthemedianfull-timewage.Butcrucially,theseotheraffluentcountriesalsoprovideamuchhigher“socialwage”thantheUnitedStatesintheformofuniversal(notmeans-tested)supportforhealth,education,andespeciallychildsupport.Inthecurrentcontext,thelegalwagefloormustcarryamuchhigherburdenformaintainingminimallydecentfamilyincomesintheUnitedStatesthaninotheraffluentcountries.Atthesametime,itisunrealistictoexpectanationalstatutoryminimumwagetocarrythisentireburden.OneapproachwouldbetocomplementaMinimumLivingWage(MLW),setperhapswithreferencetoabasic-needsbudgetforasingleadultworker,complementedbyafederaluniversaltaxablechild-cash-benefitprogramalongthelinesmodeledbyGarfinkeletal.(2016).ThiswouldbesimilartoallowancesystemsalreadyinplaceincountriesincludingCanada,theUnitedKingdom,andFrance.Garfinkeletal.showthatunconditionalcashallowanceslikethesecanbemoreeffectiveatreducingpovertythanotherkindsofpoliciessuchasachildtaxcredit.AstheprominentU.K.economistAnthonyAtkinsonhasargued,“AChildBenefitthatissubstantialbuttaxable,combinedwithaprogressive(income)ratestructure…isaneffectivewayofensuringthatallfamiliesreceivesomerecognitionoftheirfamilyresponsibilitiesbutthatmoreisgivenperchildtothoseonlowerincomes”(Atkinson,2015,p.2014).Still,therootoftheproblemofin-workpovertyisinadequatepay,andthatiswherethemainsolutionmustbefound.AstheUniversityofChicagoeconomistPaulH.Douglas(1925,p.16)wroteinremarkablystronglanguageforanacademicjournal:
Itisthemostcruelformofunconscioushypocrisyforbusinessmentopayinsufficientwagestothoseoftheiremployeeswhoarefathersoffamiliesandthenbycontributingtochildwelfareagenciestofeelthattheyhavedischargedtheirduty…employersandsocialworkersalikeneedtobewareofthinkingthatsocialwork,forallitsvaluablecontributions,isanadequatesubstituteforadecentwage.
49
Partoftheexplanationforourpersistentfailuretoestablishaminimallydecentwageflooratthefederallevelhasbeenthewaythediscoursehasbeenframed—evenbymanyofthestrongestadvocatesforsubstantiallyhigherminimumwage.Therhetorichasbeendominatedbyeconomics101thinking,inwhichincreasesintheminimumwageareconstrainedbytheeconomist’sParetoCriterionof“noharmtoanyone.”It’smanifestationinthecurrentAmericanminimumwagedebateiswhatwehavecalledtheNo-Job-Loss(NJL)criterion.Thisrhetoricisthemirror-oppositeofthemoraloutragevoicedinbyintellectualandpoliticalleadersofearliergenerations,fromleadingeconomistssuchasAdamSmithandPaulDouglas,tothepatricianpoliticiansWinstonChurchillandFranklinRoosevelt.AsTonyAtkinsonhasargued,toeffectivelycombatpovertyandinequality,weoftenneedachangeinthediscourse.Concerningthedebateovertheminimumwage,thecriterionforsettingtheappropriatelevelofthenationallegalwagefloorshouldnotbedrivenbystatisticalcontestsoverwhatparticularwagethresholdposes“littleornoriskofjobloss,”butratherbywhatwagewillensureaminimallydecentstandardoflivingfromfull-timework,andwhatpoliciescancomplementaMinimumLivingWagethatwillensurethatanycostsofjoblossareadequatelycompensated.Ifwereallycareaboutmaximizingemploymentopportunitiesthenwewouldputamuchhigherpriorityonfull-employmentfiscalandmonetarymacroeconomicpolicy,minorvariationsofwhichhadmassivelygreateremploymenteffectsthaneventhehigheststatutorywagefloorsthathavebeenproposed.ButitisalsowellwithinourcapabilitiestocounteranyjoblossthatcanbelinkedtotheadoptionofwhatJ.B.Clarkin1913called“emergencyrelief”suchasextendedunemploymentbenefits,educationandtrainingsubsidies,andpublicjobsprograms.AMinimumLivingWagecombinedwithmeaningfulchild-cashallowanceswouldputtheUnitedStatesbackamongotheraffluentnationsbypromotingworkincentiveswhileallbuteliminatingbothin-workpovertyandchildpoverty.Itwouldputthecountryintowatersthatmostotheraffluentnationshavechartedandarealreadynavigating.
50
ReferencesAllegretto,SylviaandMichaelReich.2015.AreLocalMinimumWagesAbsorbedbyPriceIncreases?EstimatesfromInternet-basedRestaurantMenus.IRLEWorkingPaper#124-15.Berkeley,CA:InstituteforResearchonLaborEmployment,UniversityofCalifornia,Berkeley.
Allegretto,Sylvia,ArindrajitDube,andMichaelReich.2011.DoMinimumWagesReallyReduceTeenEmployment?AccountingforHeterogeneityandSelectivityinStatePanelData.IndustrialRelations50(2):205-240.
Anker,Richard.2011.EstimatingaLivingWage:AMethodologicalReview.ConditionsofWorkandEmploymentSeriesNo.29.Geneva:InternationalLabourOffice.
Atkinson,Anthony.2015.InequalityWhatCanBeDone?Cambridge,MA:HarvardUniversityPress.
Belman,Dale,andPaulJ.Wolfson.2014.TheNewMinimumWageResearch.EmploymentResearch21(2):4-5.
Brown,Charles,CurtisGilroy,andAndrewKohen.1981.TheEffectoftheMinimumWageonEmploymentandUnemployment.JournalofEconomicLiterature20(2):487-528.
BureauofLaborStatistics.2015.AProfileoftheWorkingPoor2013.BLSReport1055.
BureauofLaborStatistics.1970.YouthUnemploymentandMinimumWages.Bulletin1657.Washington,DC:U.S.DepartmentofLabor,BureauofLaborStatistics.
Card,David,AlanB.Krieger.1995.MinimumWagesandEmployment:ACaseStudyoftheFast-FoodIndustryinNewJerseyandPennsylvania.TheAmericanEconomicReview84(4):772-793.
Card,David,AlanB.Krueger.2015.MythandMeasurement.TheNewEconomicsoftheMinimumWage.Twentieth-AnniversaryEdition.Princeton:PrincetonPress.
Chapman,JeffandJeffThompson.2006.TheEconomicImpactofLocalLivingWage.EPIBriefingPaper#170.Washington,DC:EconomicPolicyInstitute.
Chau,Michelle.2011.MakingWorkPayinMontana.NewYork,NY:NationalCenterforChildreninPoverty.Clark,JohnBates.1913.“TheMinimumWage.”TheAtlanticMonthly112:289-97.
Clark,Krissy.2014.MeettheWomanbehindIkea’sLivingWageCalculator.Marketplace,August6,2014.URL:
51
http://www.marketplace.org/2014/08/06/business/meet-woman-behind-ikeas-living-wage-calculator.
CongressionalResearchService.2013.TheFairLaborStandardsAct(FLSA):AnOverview.CRSReportforCongress.
Cooper,David,JohnSchmitt,andLawrenceMishel.2015.WeCanAfforda$12.00FederalMinimumWagein2020.EPIBriefingPaper#398.Washington,DC:EconomicPolicyInstitute.
Cooper,David.2016.InflationmakesProposedMinimumWageIncreasesMoreModestthantheyAppear.EPIWorkingEconomicsBlog,February26,2016.URL:http://www.epi.org/blog/inflation-makes-proposed-minimum-wage-increases-more-modest-than-they-appear/
D’Arcy,ConorandAdamCorlett.2015.TakinguptheFloor:ExploringtheImpactoftheNationalLivingWageonEmployers.London,UK:ResolutionFoundation.
Doucouliagos,Hristos,andStanley,T.D.2009.PublicationSelectionBiasinMinimum-WageResearch.AMeta-RegressionAnalysis.BritishJournalofIndustrialRelations47(2):406-428.
Douglas,PaulH.1925.TheFamilyAllowanceSystemasaProtectorofChildren.TheAnnalsoftheAmericanAcademyofPoliticalandSocialScience121:16-24.
Dube,Arindrajit,WilliamT.Lester,andMichaelReich.2010.MinimumWageEffectsAcrossStateBorders:EstimatesUsingContiguousCounties.ReviewofEconomicsandStatistics92(1):945-964.
Dube,Arindrajit.2014.Proposal13:DesigningThoughtfulMinimumWagePolicyattheStateandLocalLevels.TheHamiltonProject(TheBrookingsInstitution).
EPI–EconomicPolicyInstitute.2015.WageDeciles(personalcommunication,Nov.23,2015).
Fairris,David,DavidRunsten,CarolinaBriones,andJessicaGoodheart.2005.TheLosAngelesLivingWageOrdinance:EffectsonWorkersandEmployers.LosAngeles,CA:LosAngelesAllianceforaNewEconomy.
Fairris,David.2005.TheImpactofLivingWagesonEmployers:AControlGroupAnalysisoftheLosAngelesOrdinance.IndustrialRelations44(1):84-105.
Fass,Sarah,KinseyAldenDinan,NancyK.Cauthen,andJessicaPurmort.2008.MakingWorkPayforIowa’sFamilies.NewYork,NY:NationalCenterforChildreninPoverty.
52
FLSA–FairLaborStandardsActof1938,asamended(RevisedMay2011).WHPublication1318.URL:http://www.dol.gov/whd/regs/statutes/FairLaborStandAct.pdf
Fredericksen,Allyson.2015.PayUp:LongHoursandLowPayLeaveWorkersataLoss.Seattle,WA:AllianceforaJustSociety.
Garfinkel,Irwin,DavidHarris,JaneWaldfogelandChristopherWimer.2016.DoingMoreforOurChildren:ModelingaUniversalChildAllowanceorMoreGenerousChildTaxCredit.TheCenturyFoundation,March16,2016.
Grimshaw,Damian,JillRubery,andMatJohnson.2016.“RaisingtheUKnationalminimumwage:acaseofactivatingorarrestingsocialdialogue?,”unpublishedmanuscript.
Grossman,Jonathan.1978.FairLaborStandardsActof1938:maximumstruggleforaminimumwage.MonthlyLaborReview101(6):22.
Hirshman,AlbertO.1991.TheRhetoricofReaction.Cambridge,MA:BelknapPressofHarvardUniversityPress.
Howel,DavidandAnnaOkatenko.2010.Bywhatmeasure?AcomparisonofFrenchandUSlabormarketperformancewithnewindicatorsofemploymentadequacy.InternationalReviewofAppliedEconomics24(3):333-357.
Howes,Candace.2005.LivingWagesandRetentionofHomecareWorkersinSanFrancisco.IndustrialRelations44(1):139-63.
ILO–InternationalLabourOffice.2014.MinimumWageSystems.InternationalLabourConference,103rdSession,2014.Geneva:InternationalLabourOffice.
JeroldWaltman.2004.TheCasefortheLivingWage.NewYork:AlgoraPublishing.
Krueger,AlanB.2015.HowMuchisTooMuch?NewYorkTimes.October11,2015.
Lindert,Peter.2004.GrowingPublic.Cambridge,UK:CambridgeUniversityPress.
LowPayCommission.2014.TheFuturePathoftheNationalMinimumWage.
LowPayCommission.2016.NationalMinimumWage:LowPayCommissionReport,Spring2016.
Luce,Stephanie.2014.FightingforaLivingWage.Ithaca,NY:CornellUniversityPress.
Manning,Alan.2016.TheElusiveEmploymentEffectoftheMinimumWage.CEPDiscussionPaperNo.1428.Marchal,SarahandIveMarx.2015.“StemmingtheTide:WhatHaveEUCountriesDonetoSupportLow-WageWorkersinanEraofDownwardWagePressures?”IZADPNo.9390(September).
53
NationalCenterforChildreninPoverty.2008.MakingWorkPayforIowa’sFamilies.
NLIHC–NationalLowIncomeHousingCoalition.2015.OutofReach2015.Washington,DC:NLIHC.
Neumark,DavidandWilliamWascher.2008.MinimumWages.Cambridge,MA:MITPress.
Neumark,David,J.M.IanSalas,andWilliamWascher.2014.MoreonRecentEvidenceontheEffectsofMinimumWagesintheUnitedStates.NBERWorkingPaperNo.20619.
OECD.2006.BoostingjobsandIncomes:PolicylessonsfromreassessingtheOECDjobsstrategy.Paris:OECDPublishing.
OECD.2015.EducationataGlance2015.Paris:OECDPublishing.
OECD.2015a.MinimumWagesAftertheCrisis:MakingThemPay.
OECD.2015b.EmploymentOutlook2015.Paris:OECDPublishing.
Parsa,H.G.,JohnT.Self,DavidNjiteandTiffanyKing,2005.WhyRestaurantsFail.CornellHospitalityQuarterly46(3):304-324.
Polanyi,Karl.2001[1944].TheGreatTransformation.Boston,MA:BeaconPress.
Prasch,RobertE.2000.JohnBatesClark’sDefenseofMandatoryArbitrationandMinimumWageLegislation.JournaloftheHistoryofEconomicThought22(2):251-263.
Reich,Michael,PeterHallandKenJacobs.2005.LivingWagePoliciesattheSanFranciscoAirport:ImpactsonWorkersandBusinesses.IndustrialRelations44(1):106-138.
Reich,Michael,PeterHall,andKenJacobs.2005.LivingWagePoliciesattheSanFranciscoAirport:ImpactsonWorkersandBusinesses.IndustrialRelations44(1):106-138.
Reich,Michael,SylviaAllegretto,KenJacobsandClaireMontialoux,2016.TheEffectsofa$15MinimumWageinNewYorkState.Berkeley,CA:CenteronWageandEmploymentDynamics,InstituteforResearchonLaborandEmploymentUniversityofCalifornia,Berkeley.
ResolutionFoundation.2014.MorethanaMinimum.TheResolutionFoundationReviewoftheFutureoftheNationalMinimumWage:TheFinalReport.
Roosevelt.FranklinD.1937.MessagetoCongressonEstablishingMinimumWagesandMaximumHours(May24,1937).OnlinebyGerhardPetersandJohnT.
54
Woolley,TheAmericanPresidencyProject.http://www.presidency.ucsb.edu/ws/?pid=15405.
Ryan,J.(1912).Alivingwage:Itsethicalandeconomicaspects.London:MacMillan.
Scheiber,Noam.2015.RaisingFloorforWagesPushesEconomyintotheUnknown.NewYorkTimes.July27,2015.
Schmitt,John.2013.WhyDoestheMinimumWageHaveNoDiscernibleEffectonEmployment?CEPRReport.Washington,DC:CenterforEconomicandPolicyResearch.
Schmitt,John.2015.ExplainingtheSmallEmploymentEffectsoftheMinimumWageintheUnitedStates.IndustrialRelations54(4):547-581.
Smith,Adam.[1776]1937.TheWealthofNations.NewYork,NY:TheModernLibrary.
Stabile,Donald.2008.TheLivingWage:Lessonsfromthehistoryofeconomicthought.Cheltenham:EdwardElgarPublishing.
Tung,Irene,YannetLathrop,andPaulSonn.2015.TheGrowingMovementfor$15.NELPReport.
U.S.DepartmentofLabor,BureauofLaborStatistics.LaborForceStatisticsfromtheCurrentPopulationSurvey:Monthlyteenunemployment(seriesLNS14000012).
Webb,Sidney.1912.TheMinimumWage.JournalofPoliticalEconomy20(10):973-998.
Welch,FinisR.1995.CommentsbytheReviewers.IndustrialandLaborRelationsReview48(4):828-849.
Werner,AndreaandMingLim.2015.Theethicsofthelivingwage:areviewandresearchagenda.JournalofBusinessEthics.Publishedonline:February12,2015.
Wicks-Lim,JeannetteandRobertPollin.2012MakingWorkPay:CombiningtheBenefitsoftheEarnedIncomeTaxCreditandMinimumWage.Amherst,MA:PoliticalEconomyResearchInstitute.
Wofford,Ben,andManuelaTobia.2016.ASurprisingNumberofAmerica’sMayorsSupport$15MinimumWage.Politico.January25,2016.
55
Appendix:AKaitzIndexThoughtExperimentArinDube(2014)hasproposedthatthecriterionforsettingthestatutoryminimumwageshouldbetheKaitzindex—theratiooftheminimumtotheaverageormedianwage—andfurtherproposesthatthe“natural”and“appropriate”Kaitzratioshouldbe50percent.ThiswouldraisetheminimumwageeverywhereintheUnitedStates,butbasedonhisassessmentoftheevidence,employmenteffectswould“likely(be)toosmalltobemeaningfullydifferentfromzero”(p.8).Thisisagoodexampleoftheapplicationofthebackward-lookingNo-Job-Loss(NJL)criterion:theminimumwageshouldbesetatthehighestwagethatevidenceshowswillposelittleornoriskofjobloss.Asheputsit,
…acomparisonoftheminimumwagetothemedianoffersaguideforhowbindingaparticularminimumwageincreaseislikelytobe,andwhattypeofwagethelabormarketcanbear.Whenthisratioislow—sayaround0.2—minimumwagepolicyisnotraisingthewagesofmanyworkers.Incontrast,ahighratio—sayaround.08—indicatesahighlyinterventionistpolicywheretheminimumwageisdramaticallycompressingdifferencesinwagesfornearlyhalftheworkforce….Nooneexpectsthattheminimumwageshouldbesetequaltothemedianwage….(p.2).
ThatconclusionmaybeafairdepictionofthemainstreamU.S.minimumwagediscourse,butmanycountrieshavechosenpoliciesthatseverelycompressthebottomofthewagedistributionandhavedonesoexplicitlyonethicalcriteriaofwhatisaminimallydecentincomefromwork.The10-50(or50-10)ratioisastandardmeasureofinequalityatthebottomofthewagedistribution.Themostrecentdata(2013-15)showthattheratioofgrossearningsofthe10thpercentileworkertothemedianworkerwas45.5percentfortheUnitedStates(notmuchabovetheU.S.Kaitzratioof37percent).Thiscomparesto55.5percentand57.5percentfortheUnitedKingdomandAustralia,respectively,justover68percentforDenmarkandFinland,and72percentforBelgiumandSweden.Ahighlycompressedlow-endwagedistributionisclearlycompatiblewithahigh-employmentlabormarketintheaffluentworld.ThefundamentalproblemwiththeKaitzindexasaguidetotheriskofjoblossisthatthelevelandchangeofalocation’smedianwage—thedenominatoroftheratio—mayhavelittleornothingtodowiththedynamicsofwageandemploymentsettingatthelevelofthefirm.Dube’sproposalfocusesontwolevelsatwhichtheKaitzcanbeoperative,statesandmetropolitanareas.IftheKaitzindexdefinedfor
56
geographicjurisdictionsisatallusefulasaguidetothelikelihoodofconsequentialjobloss,itshouldperformbestatthelocallevel.Thefollowingthought-experiment(withfairlyrealisticnumbers)showsthattheKaitzratiowillbewildlydifferentforthesameworkersemployedinthesamefastfoodfranchisesdependingonthejurisdictionforwhichtheKaitzratioiscalculated,rangingfromthetwoextremes(.8and.2)intheDubepassagequotedabove.Webeginwithanextremelylocallabormarket:let’scallitabigreststopontheNewYorkThruwaythatisthedominantemployerinthearea,whichisrural.Let’ssaythereare300employedworkers,rangingfrommanagerstoentry-levelcashiers,cleaners,andmaintenancestaff.AssumethattheNewYorkstateminimumwagehasrecentlyrisento$9,pushingupthewagesofmanyoftheworkersandseverelycompressingthebottomhalfofthewagedistribution.Asaconsequence,half(150)arenowpaidlessthan$11,sothemedianwageinthislabormarketis$11,andtheKaitzindexis82percent($9/$11).SincethisreststopislocatedinupstateNewYork(abovethenorthernNewYorkCitysuburbs)whichisscheduledforaslowerphase-inofahigherwagefloor(perhapsto$15),thisisanotherrelevantjurisdictionforwhichtheKaitzindexcanbecalculated.Iftherelevantlabormarketisthisupstateregionandthemedianwageis$18,thentheKaitzindexwillbe50percent($9/$18).Butiftheentirestateistherelevantjurisdictionandthemedianis$27,theoperativeKaitzratiois33percent($9/$27).Thisexampleillustrateshowdifferencesinmedianwagesacrossdifferentpoliticaljurisdictionscancausethe$9state-widewagefloortogenerateKaitzindexvaluesrangingfrom33to82percent,dependingonwhetherthereferencemedianwageisdefinedatthelevelofthereststop,upstateNewYork,ortheentirestate.ButforeachKaitzvalue,theworkersarethesame,thelikelihoodthattheworkerquits(turnover)isthesame,theabilitytopayforpastandfutureminimumwageincreasesbyreducingpayraisesforthetopearning150workersisthesame,andtheabilitytopassalonglaborcostincreasesinpricesorreducedprofitsisthesame.Whatmakesthe50percentKaitzratiotherightone?WhatwouldtheKaitzindexcalculatedforeachofthesejurisdictionstellusaboutthelikelyemploymenteffectsofanotherminimumwagehike,sayto$10(asthephase-intothe$15wagebegins)?Ifthisincreaseinthestatewagefloorpushedthemedianwageatthereststopto$12buthadnoeffectontheupstateorstate-widemedians(tokeeptheexamplesimple),thenthechangesintheKaitzratioswould
57
be:from82percent($9/$11)to83percent($10/$12)forthereststoplabormarket;from50percent($9/$18)to56percent($10/$18)fortheupstatelabormarket;and33percent($9/$27)to44percent($10/$27)forthestate-widelabormarket.Whichofthesechangesshouldbeusedasthebestguideto“whattypeofwagethelabormarketcanbear”?Turningthisthoughtexperimentaround,wecanaskabouttheimplicationsforthe“natural”wagefloorforeachjurisdiction,shoulda50-percentrulebeusedtosettheminimumwageafterthestatutorywagefloorisincreasedfrom$9to$10.Ifthelabormarketwasheldtobethelocalarea—overwhelminglydominatedbythe300workerreststop—thenthetheappropriatewagewouldbejust$6(50percentof$12).IfmanyupstateNewYorkStatelegislatorsarerightthattheupstateareaisthemostappropriatelabormarketforthepurposesofdeterminingminimumwageeffects,thenthe50percentKaitzrulewouldgenerateawagefloorof$9(50percentof$18).Butifthosewhoarguethattheminimumwageshouldbesetfortheentirestate,theproperwagefloorwouldbe$13.50(50percentof$27).Doesaformulathatgeneratesoutcomesthatrangefrom$6to$13.50offerausefulguidetosettingtheappropriateminimumwageforourNewYorkThruwayreststopworkersandtheiremployers?
Related Documents
