Treść książki
Przejdź do opcji czytnikaPrzejdź do nawigacjiPrzejdź do informacjiPrzejdź do stopki
ShanshanLiu,ThomasHyclak
economists.Previousstudies,boththeoretical
panelmodelsisthatspecifiedinNeumarkand
andempirical,havegenerallyreachedavarietyof
Wascher(1992)andwritteninequation(1).Here
conclusionsabouttheeffectsontheemployment
E
it
representsemploymentinstateiattimet,
level.Thelargeempiricalliterature,mainlylook-
MWitistheminimumwageinstateiattimet
ingatteenageworkersorthoseemployedin
andXitincludescontrolvariables.Theequation
restaurantsorotherestablishmentslikelytobe
alsoincludesyeardummiesτ
t
tocontrolfor
affectedbytheminimumwage,canbedivided
state-invarianttimeeffectsandstatedummiesϕ
i
to
intofourclusters,orderedchronologically.
capturestate-specific,time-invariantunobserved
First,bytheearly1980s,moststudiesofthe
characteristics:
effectsoftimeseriesvariationinthenational
minimumwagereportedresultssuggestingthat
Eit=β0+β1MWit+Xitβ2+ϕi+τt+εit,
(1)
higherminimumwagesdecreasedemployment
opportunitiesforlowwageworkers.Brown’s
NeumarkandWascher(2008)concludethat
influentialsurvey(1982)identifiedaconsensus
panelstudiesfocusingoncross-statevariation
findingthattheminimumwageelasticityof
withtimeandstatefixedeffectssuggestrenewed
teenageemploymentrangedfrom-0.1to-0.3
supportforthe“consensus”employmentelasticity
intimeseriesstudiesusingCurrentPopulation
estimateof-0.1to-0.3.
Surveydatawithvaryingsampleperiodsand
Severalrecentpapersoftheemploymenteffects
specifications.
ofcross-statevariationinminimumwagelevels
Second,startinginthe1990s,theresultsofthe
reinforcetheconclusionofthisthirdcluster
“newminimumwageresearch”1,whichtended
ofstudies.Thompson(2009),usingQuarterly
tostudycross-statedifferences,raisedquestions
WorkforceIndicatorsdatafor1996-2000,evalua-
abouttheeffectsonlow-wageemploymentof
teshowstatedifferencesinminimumwages
ahigherminimumwage(NeumarkandWascher,
affectteenageemploymentatthecountylevel.
2007).Casestudiesofcrossborderdifferences
Byreferringtoquintilesofteenaveragequarterly
followinganincreaseinastate’sminimumwage
earnings,heidentifieshigh-impactcounties
level(CardandKrueger1994and1995)oftenfind
(wheretheminimumwageismostlikelytoaf-
apositiveandstatisticallysignificanteffectofthe
fectteenagemarkets)andlow-impactcounties
minimumwageonemploymentinlow-wagelabor
(whereprevailingwagesexceedtheminimum).
markets.Thiseffectcanbeexplained,inpart,
Usingdifference-in-differenceestimations,he
byassumingthatemployersoflow-wageworkers
demonstratesthattheemploymentelasticityin
havemarketpowerandactasmonopsonistic
thehigh-impactcountiesrangedbetween-0.26
buyersoflabor.
and-0.37.Inaddition,usingthesameDIDmodel
Third,panelstudies,employingnational-level
withanalternativedependentvariable(hiring),
longitudinaldataonindividualsortime-series
heshowsthattheteenshareofnewhires(THS)
dataforacross-sectionofgeographicareas,often
declinedmarkedlyfollowingaminimumwage
findanegativecorrelationbetweenemployment
increase.
andtheminimumwage.
2
Agoodexampleofsuch
Sabia(2009)pointsoutthatindustrystudies
narrowlybasedonsub-sectorsoftheretailsector,
1ThenewminimumwageresearchbeganinNovember
suchasthefast-foodrestaurantsinmanycase
1991,whentherewereaninnovativesetofstudiesonthe
studies,maynotcapturetheminimumwageef-
effectsoftheminimumwagepresentedanddiscussedin
fectsacrosstheentireretailsector.Usingmonthly
“NewMinimumWageResearchConference”.Andaspe-
datafromthe1979-2004CPS,Sabiasupplements
cialissueoftheIndustrialandLaborRelationsReview
(ILRR)waspublishedintheearly1990s.(Neumarkand
Wascher,2007)
bothstateandtimevariationinminimumwages.For
2Generallyspeaking,nationwideindividual-leveldata,
example,QuarterlyCensusofEmploymentandWages
suchastheCurrentPopulationSurvey(CPS),canprovide
(QCEW)andQuarterlyWorkforceIndicators(QWI)can
theworker-leveldemographicinformationtoestimatethe
provideafullcensusofquarterlycountofemployment
employmenteffectbyage,genderorrace;time-seriesand
andwages,availableatthecounty,MSA,andstatelevels
cross-sectiondata(thegeographicdata)canincorporate
byindustry.
6
