[PDF] Gender Gaps in Early Wage Expectations

View - Download Gender Gaps in Early Wage Expectations

Previous PDF

Next PDF

DISCUSSION PAPER SERIES

IZA DP No. 15281

Andreas Leibing

Frauke Peter

Sevrin Waights

C. Katharina Spiess

Gender Gaps in Early Wage Expectations

MAY 2022

Any opinions expressed in this paper are those of the author(s) and not those of IZA. Research published in this series may

include views on policy, but IZA takes no institutional policy positions. The IZA research network is committed to the IZA

Guiding Principles of Research Integrity.

The IZA Institute of Labor Economics is an independent economic research institute that conducts research in labor economics

and offers evidence-based policy advice on labor market issues. Supported by the Deutsche Post Foundation, IZA runs the

world's largest network of economists, whose research aims to provide answers to the global labor market challenges of our

time. Our key objective is to build bridges between academic research, policymakers and society.

IZA Discussion Papers often represent preliminary work and are circulated to encourage discussion. Citation of such a paper

should account for its provisional character. A revised version may be available directly from the author.

Schaumburg-Lippe-Straße 5-9

53113 Bonn, Germany

Phone: +49-228-3894-0

Email: publications@iza.orgwww.iza.org

IZA - Institute of Labor Economics

DISCUSSION PAPER SERIES

IZA DP No. 15281

Gender Gaps in Early Wage Expectations

MAY 2022

Andreas Leibing

Frauke Peter

DZHW Hannover

Sevrin Waights

C. Katharina Spiess

Federal Institute for Population Research, Johannes Gutenberg University Mainz and IZA

ABSTRACT

IZA DP No. 15281MAY 2022

Gender Gaps in Early Wage Expectations

Using detailed data from a unique survey of high school graduates in Germany, we document a gender gap in expected full-time earnings of more than 15%. We apply a regression-compatible Oaxaca-Blinder decomposition and find that especially differences in coefficients help explain the gap. In particular, the effects of having time for family as career motive and being first-generation college student are associated with large penalties in female wage expectations exclusively. This is especially true for higher expected career paths. Resulting expected returns to education are associated with college enrollment of women and could thus entrench subsequent gaps in realized earnings.

JEL Classification: I26, J31, D84

Keywords: wage expectations, gender gap, college enrollment

Corresponding author:

C. Katharina Spiess

Friedrich-Ebert-Allee 4

65185 Wiesbaden

Germany

E-mail: C.Katharina.Spiess@bib.bund.de

* We thank Teodora Boneva, Gabriella Conti, Thomas Dohmen, Bernd Fitzenberger, Henning Hermes, Hannah

of the annual meeting of the German Ecomomic Association, the meeting of its standing field committee in Education

Economics and of the 1

st Berlin Workshop on Empirical Public Economics for valuable comments and suggestions. We

especially thank Vaishali Zambre for her initial work on wage expectations in the Berliner-Studienberechtigten-Panel.

Frauke Peter and C. Katharina Spiess gratefully acknowledge funding from the Einstein Foundation Berlin (A-2010-

025 (FU)). Andreas Leibing, Frauke Peter, Sevrin Waights and C. Katharina Spiess further acknowledge funding from

the German Science Foundation (SP 1091/2-1). The usual disclaimer applies.

1Introduction

Thelitera tureongenderpaygapscontinue stogro wrapidlyinparallelwit halively publicdeba te(seeBlauandKahn ,2017,fo ranovervie w).Man ystudiesshowthatgender gapsarenot onlyprevalent inactual earnings ,butthatalreadyincollege,women hold lowerwageexp ectationsthanmen (e.g.BlauandFer ber,1991,Filippinand Ichino,2005, Reubenetal.,2017).Howe ver,littleisknownaboutwagee xpectatio nsof individuals beforetheirdecisiontopurs uetertiary education.Suchearlywageexpectat ionscaninflu- encelaterhuma ncapitalinvestme ntandcare erdecisions.Hence,abe tterunderstanding ofearly gendergapsinexpe ctedwagesca nalsoprovideins ight sintotheformationof laterdisparities. Gendergapsinexp ectedwage smay entrenchactualwageinequalitiest hroughat leasttwochannels .First,based onhumancapitaltheor y,lowerexpect edwagesreduce theincent ivestoinvestineducation.Exis tingevidenc eshowsthatexpect edw agesare asig nificantpredictorforthecho icewhichlevelofeducation topursue,asw ellasfo r collegemajorchoice. 1 Second,wageexpe ctationspriortola bormarketentrycanbecome self-fulfilling(DeLaR icaetal. ,2008).For instance,wag eexpectationsmighta!ect startingwagesthroughthe formationofreser vationwages(BrownandTaylor,2013).If, intur n,lowerrese rvationwages resultinlowerstartingsalaries,theyarelik elytohave ape rsistente!ectonfuture wagetr ajectories. 2

Caliendoetal.(2017)s howthatgender

di!erencesinreservation wag escanindeedaccountforalargeshareoft hesubsequent gendergapinrealiz edwages. Thisis exacerbatedbygenderdi!erencesinnegotiation stylesandoutcomes, whichare largerforinexperiencednego tiators(se eMazeietal.,

2015,fo ranoverv iew)andt husmatterespeciallyforstarting wages.

Ourstudya sks:whatfa ctorsdrivethegender gapinexpec tedwagesamonghigh schoolstudents?Oura nalysisdrawsondatafromaunique surveya monghighschool graduatesinGermany,inwhichw eas kedindividualstudentsfortheirex pec tedfull-time earningsrangeatage3 5yearsinthree di!erentscenarios:(i) iftheyenrolledforavo- cationaldegree,(ii)ifthey enrolledforabachelor'sdeg reeor( iii)ift heyenrolledf or amas ter'sdegree.Weexaminet hedeterminantsofthegender gapusinga regression- compatibleOaxaca-Blinde r(OB)decomposition(Oaxaca,1973,Blinder,1973,Fortin,

2008)and placeapar ticularfocuso ndi!erencesincoe"cients(unexplainedgap)as

opposedtodi!erencesinendowments(explainedgap),givinga detailedover view ofcom- ponentsattributabletoso cio-demographicfactors,inte ndedcolleg emajorchoice,career motivesandbothcognit iveandno ncognitiveabilities. 1 Seee.g.AttanasioandKaufmann(2014,2017),Belfieldetal.(2020)for highsch oolandcollege enrollment andBonevaetal.(2021)for postgradu ateenrollment.Forcollegemajor choiceseeZafar(2013),Ruder andNoy(2017),WiswallandZafar(2018)and Arcidiaconoetal.(2020). 2 Evidenceontheadversee!ectoflower star tingwagesisprovi dedbyOreopoulosetal.(2012).The authorsshowthatent eringthelabor marketduri ngarecessionhaspotentiallyl onglastingscarring e!ectsonwage s. 1 Wefindthat theg endergapinav erage wageexpectat ionsafterhigh sc hoolgraduation amountstoover15%.O urdecompo sitionresultsindica tethatendowme ntsex plainsome ofthedi!erenceinfuturewages :ar oundaquarterof thetotalgap.Inparticular ,inten- tionstochoo seamajo rinScience,Technology ,Eng ineering,andMathema tics (STEM) andahigh aca demicself- e"cacy 3 aremoreprev alentinmenand onaverageassociated withhigherexp ectedwage s.However,itistheunexplainedcoe"cientsratherthanen- dowmentsthatplayabiggerro le.Beingpot entialfir stg enerationatco llege,intendingto studybusinesso rmanagementandex pressingtimeforfamily asacar eer motiveallhave significantlymorenegativeimpactso nfemaleexp ectedwagesthanonmale expected wages.Somefurtherfacto rs,suc hashavinggoodhealthandsafetyconditions asac areer motive,alsohaverela tivelymorepo sitiveimpactso nfemaleexpectations.Thefa ctthat thesecoe"ciente!ectso!seteachot her,explainstherela tivelylowshareofthe gender gapinwagee xpe ctationsexplained bydi!erencesincoe"cients. Weinves tigatethedecompositionnotjustformea nexpec tedwagespooledovereduca- tionalscenarios, butalsoforminimumandmaximumexpected wag es,aswella sbachelor andmaste rscenariosseparate ly,toshedlightonunderlyingheter ogeneity.Certainfac- tors,e.g.timeforfamily ,ar eassociate dwiththelargestunexplainedcoe"cientsforthe maximumexpecte dwagesandthemasterscenario,implying thedi!erentialimpactsare exacerbatedforthehighestcareertrac ks.Thisresult suggest sthatwomenant icipate havingtogiveup highercar eerpathsand leade rshippositions inordertohavemore flexibleworka rrangements. 4

RecentevidencebyWiswallandZafar(2021)st ressesthe

closelinkbetween human capitalinvestments,includingmajo rchoice,and expectations aboutcareerand family. Inourfinding s,havinga veryhighpref erencefor timeforfamily asacar eer motive isthe largest singleandmostconsistentdr iverofthege ndergapin wageexpe ctations. Weshow thatitiscompar ableinsizetot heac tua lmotherhoodpenaltyinfull-time ear n- ingsofwo rkers thatholdatleastabachelor'sdeg reearoundthe ageof3 5inGerma ny. Thiscompar isonsuggeststhatthemotherho odpenalty,thebiggestfacto rinexplaining earningsinequality,isalre adypresentintheexpect atio nsofhighschoolstuden ts.Fi- nally,inordert oas sesspossiblepat hwayso fhowthegendergapinwage expectations mighttransla teintoactualearnings,wealso examinetheas sociationbetweene xpected returnstocollegeandc ollegeenr ollment.Wefindstronghe teroge neitiesacrossgender andsocio economicbackground. Themajor ityofexistingstudiesonea rningsexp ectationsfocusona verages,asking studentsforpointestimat esoftheirf uturewages(for overviews,seee.g.Brunelloetal.,

2004,Manski,2004).Byelic itingtheminimum andmaximumexpec tedwag esofhigh

3 Academicself-e"cacyrefers toindividualsconvic tionsth attheycansuccessfullyperformgiven academic tasksatdesignat edl evels(BongandSkaal vik,2003).See Section2fordetai ls. 4 Thisphenomen oniswellreflectedintherel atively lowshareofwomen inleadingpositionsinfirms (KirschandWrohlich,2020)bu talsoinacad emia(CeciandWilliam s,2011). 2 schoolstudents,wea reabletogiveamorecomprehensiv epicturea ndareableto assess driversofwageexpe ctatio nsatdi!erentmargins.Toas sesstheaccuracyof expec tations, somestudiesint hisstrandofthelit era turecompa reexpectedwage sofstude ntsto actualwagesofdi!erentreferencegr oups(Manski,1993,Betts,1996,Huntington-Klein,

2015).Bye licitingwag eexpectationsf ordi!erentdegreetype s,wecancomparewag e

expectationstoactualearningsofawider rangeof referencegroups.Ot herstudies investigatedtheaccuracyofexpectat ionsbyco mparingthemtolaterrealizedwa gesand followedstudentsintothela bormarket(WebbinkandHarto g,2004,Jerrim,2015). Sofa r,onlytwostudiesc overwa geexpect ationsofhighschoo lstudents.Seminal workbyDominitzandManski(1996)as ksasmallsampleof3 9co llegestudent sand

71highsc hoolstude ntsabouttheirmediane xpectedearningsatages 30and40,condi-

tionalondi!erentschoolingscena riosandalsoelicittheirb eliefsaboutactualearnings distributions.Theauthorsfindthatbo thgender sexpectpositivere tur nstoschooling andanincr easeine arningsbetweenage30a nd40 ,whileoverestimatingc urrentearnings inequality.Thoughnotthefoc usoftheirstudy,the authors alsofindag endergapin earningsexpectations acrossschoolingscenariosandatbothage s.Furthermore,Boneva etal.(2022)elic itthegenderga pincompe titivenessofado lescentsfr omlo wersocio- economicbackgroundsinG ermanybylookingatearningsex pec tations.Theirfocusis onthede velopment ofgenderdi!erencesincompetitivenes sand theroleofthesocialen- vironmentinthisprocess.A dditio nally,thea uthorsshowthatalsoearningsex pectations ofgirlsa ndboysalrea dydi!ersignific antlyattheageoffourteen/fif teen.T heirfindings suggestthatthegenderga pinwageexp ectat ionsemergesalrea dyatthebeg inningof adolescenceandislargerforchildren fromlow ers ocio-economicbackgr ounds. Kiesslingetal.(2019)us edatafr omtheonlinesurv eyFachkraft2030thatelicits wageexpecta tionsofcollegestudentsinGermany. 5

Theirfindingsattr ibutealarg errole

inex plainingthegendergapin expec tedwagestooc cupationalsor ting andnegotiation styles.Surprisingly,prospec tivechild-relatedlaborfo rceinterruptionshavearelatively smalle!ectintheirf rame work,whichf ocusesprimarilyonendowmente!ects.Another relatedstudybyFernandesetal.(2021)use sdataonwa geexpecta tionsofs tudents majoringinbusinessattwo Swiss colleges.Theirfinding sconfirmt heimportance of bothfertilityc onsiderationsandcareerpre ferencesinexplainingthegendergapinwage expectations,butfocusmostlyonendowmente!ects.Consistentwit hourresults,they alsofindthatb oth malesandfemale soverestimate theirwagesc omparedt oactualones. Furthermore,malesrespondinanoverc onfidentmannerto informationaboutrealized wages. 5 Thedatawas collectedam ongparti cipantsoflargestjobboardjobmensa.deinGer many.Ehrmantraut etal.(2020)us ethesamed atatostudy theexpec tedsignalli ngvalue ofcompletinghighe reducation, usingcollegedropout ascounterfactual. 3 Lastly,relatedstudiesby e.g.WiswallandZafar(2018)and Reubenetal.(2017) investigatetherelationshipbetweenj obpref erences,majorchoiceandwage expectations amongcollege students,usingdatafro mhigh-abilityundergraduatestudents atNewYo rk University.WiswallandZafar(2018)co nfrontsthesestudentswithmultipleh ypothetical jobchoicesc enariosthatvary inexpectedwagesandother jobcharac teristics.Their resultsshowthat,amo ngothers,wo menhaveasignifican tlyhigherwillingnes stopay (interms ofexpectedwa ges)f orflexibleworkinghoursandmoresecure jobsthanmen. Sincetimeforfamily canbeinter preteda saformofhigherflexibilityinworkinghours, ourfindingsar eroughly inlinewiththese results.Howeve r,w eca nnotconfirmwome n's highwillingnes stopayforsecurej obs. 6

Reubenetal.(2017)als oinvestigat estherole

ofprefe rencesinexplaininggenderdi!erencesinwageexpecta tio ns.Basedonthes ame surveyamongNewYork Universityunderg raduate stheydocumentalar gegendergap inex pectedwages.Whilepartofthegap isduetogenderdi!erencesincollegemajor choice,thegapinwage expec tationswithina colle gemajorstillamoun tstoaround

20%.Theyfurt hershowthatg enderdi!erencesinpreferencessuc ha soverconfidence,

competitivenessandriskaversion,explain18%of thegende rgapinexpe ctations. Ourstudy di!ersfromthe before-mentione dstudies inthreecentralaspects.First, ourstudyis targeted atr ecenthighschoolgraduatesandtr acksthe moverthecourse oftheirp otentialunive rsitystudies.Thus,weareabletos hedlightontheimportance ofwage expectationsfo rawiderrangeofcareerdecisions,suchas collegeenrollme nt andcompa rewageexpectatio nstoaactualea rningsofawiderrangeofreferencegroups. Second,ourdataov ersamples graduateswit hnon-collegeparentalbackground.O urstudy ishenc ebasedona potentiallymorer eprese nta tivesamplethan(elite)universitystude nts (incert ainmajors).Third,ourdata allowsustogobeyondtheavera geandelicits the fullrange ofexpectedwa gesby degreetypeforeachstudent.T akentogether,ours tudy contributestoamoregeneralunders tandingofw hatf actorsdete rminewageexpectations atanea rlysta geandshedslighton theirimplications. Thepaper isorganizedasfo llow s.Section2introducesthedata,providesdet ailed informationonthemeasurementof wagee xpecta tionsandpresentsdescriptivestat istics. Section3introducestheOBdecompositionme thod.Section4presentsthedecomposition results,examiningtheroleofdi !erentfactorsine xplainingthegendergapinwage expectations.Section5showsassociat ionsofwageexpectationswithsubseque ntcolleg e enrollment.Section6concludes. 6 Ali kelyexplanationisthe factthattheGermanlabormarketi smuchm oreregulated thantheAme rican labormarket .Forexample,at-willemploymentisnotpos siblei nGermany. 4

2Dataanddescript ivesta tistics

Thedatat hatweuseinour empiricalanaly sisareba sedona surv eyofhig hschoolstu- dentsatdi!erentinstitutionsinthe GermancapitalcityofBe rlin.T heyincludedet ailed informationonstudentcharact eristics ,educationalaspiratio nsandeventualeducational choices.Alloftheseaspec ts weresur veyedaspartof alargerresearchproject ,the Berliner-Studienberechtigten-Panel(BestUp),whichaimsto studyeduca tionalpaths ofhighsc hoolstude nts(seeEhlertetal.,2017b,fo rdetails). Thesurv eyaimedatobtainingasample ofstudents whowere predominantlyfrom lowersocioecono micbackgrounds.Thus,the27selectedsc hoolsarelocatedindistricts withahighsha reo findividualswit houtcollegedegree andcov er20%ofall upper- secondaryschoolsinBerlin.Thesur veyfollowedallstude nts fromtheendoft heir penultimateyearinhighschoolthr oughtwoyears aft ergradua tingfromhighschool.All studentsweresurveyedfiv etimesoverthat period.Exceptforthefirstpap erandpe ncil surveyconductedinsc hools,thesubsequentsurve ysw ereadministeredonline.Ofthe

1,578studentss urveyedinthefirstwave, 1,105participatedintheseconda nd1 ,033in

thethirdw ave,wherew ageexpectationsw eresurveyed(s eeAppendixFigureF.1). Wageexpectat ionsinthiscontexthavefirstbeenstudiedbyZambre(2018).Those participantswhodecidedtoenrollatc ollegewe resurveyedinsubs equent wavestostudy thetrans itionintopostgraduateeducat ion.Thissoca lledPostGrad-BestUpprojectthus focusedonlyonstudents,w howere enrolledin collegein2017(seeBerkesetal.(2022) andAppendix FigureF.1).Other elementsoftheBe stUpprojectconsistedin separat e randomizedcontrolledtrials .First:aninformationinterventio nonthereturnstote rtiary educationtostudye!ectsoncollege enrollment intentions(PeterandZambre,2017),col- legeapplications (Ehlertetal.,2017a)and actual enrollment(Peteretal.,2021).Second: afinanc ialintervention informofatemporarymonthlysubsidyforstude nt swithouten- rollmentintentionstos tudythee!ectoncolleg eapplicatio ns,whichremainedine!ective (Peteretal.,2017).Eve nthoughtheseinte rventionsareno tthefocusofoura nalysis, theymightstill a!ectwagee xpectationsdi!erentlyforbothgender sandthusalso the gendergapinexp ectations(s eeApp endixFigureF.2fortheinforma tiontre atment). Therefore,weincludeindicatorvariablesac countingf orschoo l-levelassignmentto each interventionthroughoutthisstudy.

2.1Wageexpectati ons

Intheex pected earningsmoduleofthesurveyinsummer201 4,studentswereas ked tostat etheminimumandmaximumnetwa gethat theymight expect toearnatthe ageof35c onditionalo nworking fulltime. 7

FollowingGuisoetal.(2002)and Attanasio

7 Giventhatwecondi tiononfullt imeem ployment,biasesarisingf rom di!erentlaborsupply expectation s arerule doutbyconstruction .Theor iginalqu estionsuggestedaregularearnedincomearisi ngfrom anem ploymentagreementandwethereforeus ethetermswageandearningsinterchangeably.Itwas 5 andKaufma nn(2014,2017),studen tswerethenaskedwhat theythinkistheprobabilit y theywillearn morethan themidpointofther angebetwee ntheirstate dminimumand maximumandweass umeasinglet riangulardis tributionbetwee nthe minimumand maximumexpecte dwage,scalingeachhalfofthetriang lesuchtheareaofthe righthalf matchesthereporte dprobability toearnabovethemidpoint(seeAppe ndixFigureB.1). Assuc htheaverag eexpec tedwageisthencomputedasE(y)= 1!p 3 (2·y min +y max p 3 (y min +2·y max Weask edforthesethreepiec esofinforma tionforthreedi!erenteducationalsce narios, inwhic htheyhaveea rned:i)avoca tionaldegre e,ii)abachelor'sdegree ,o riii)a master's degree.Notethatouranaly sis,wefocus onbachelor's andmaste r'sdegrees. 8

Henceforth,

theterm pooledreferstoanaverageb etwee ntheset wo.Furthermore,wecanonlyc onsider individualsthatgav ewageexpe ctationsforat leastonedegree. TableA.1showsthat thisisthec as efor64 9students.Weex ploreitemnon- responsefo rthewageexpectation questioninAppendixSectionA.Whileno n-re sponseisrelatedtocertaincharac ter istics suchasaca demicperf ormance,theselectio nissimilarbetweenmenandwomen,which isimpo rtantforourstudy.Wealsoe xcludestudent swhoser esponsessugges tthatt hey didnot entirely understoodthequestion.The searestudentswhoassignedt heentire probabilitymasseithertot helowerort heupperpartoftheir individualwag edistribut ion, i.e.tothe leftort herighto fthemidpoint, aswella sstudent swhogaveeither nosuch probabilityornominimumormaxim umexpe cte dearnings suchthat thecomputation ofanav eragee xpectedwageisnotpossible. Toensurethattheanalysisisnotdriven byoutlier s,wefurtherexcludestuden tswitha naveragewageex pectationeither above thehighest orbelowthelowes tperc entileofthecros s-sect ionaldistributionofexp ected wagesforeacheduca tiondegree .Further,wedrop studentswhodidnotprovideco mplete informationonallcovariates. Thefinals amplefort heempiricalanalysisconsistsof513 students,ofwhom205aremalea nd308a refemale. Individualaverage expectedwages,pooledo verdegreesarepresentedinFigure1. 9 Inlinewit hpre viousfindingsinthelit eraturewefindalarge genderga p.Ev enbefo re enteringthelabormarketw omenex pecttoearnco nsiderablylessthantheir malecoun- terparts.Whilemenexpecttoearn ona verage around3,192EURpermonth, women expecttoearn2,691EUR .Thisdi!erenceimpliesagendergap inwagee xp ectationsof formulatedasfollows:"Nowassumet hatyouearnedavocat ionalorunivers itydegreea ndworkfull-time, meaningthatyouareecon omicallyacti veforabout 39hoursp erweek.Whatdoyouexpectyourmonthly minimumnetearningsandm onthl ymaximumnetearnings(i.e. theminim um/maximumamountof moneythatistr ansferreddir ectlyon yourbankaccount)attheageof35to beifyou...[haveearned a vocational/Bachelor/Masterdegree]". 8 Weonlyu seexpect ationscondit ionalonvocationaldegreesasreferencewh encomputingtheexpected returnstoacollegedegreet hatweu set oanalyzecollegeenrol lmentinSection5. 9 InAppen dixSectionCwealsop resentde scriptivestatisticsby genderfortheminimumandmaximum expectedwages.Di!erencesintheaveragearedr ivenm ostlybymale studentsreporti nghighe rmaxi- mumsandfemales tudent sreportinglower minimums,althoughthisdepen dssomewhatonthelevelof degree. 6 Figure1:Expe ctedwa gesbyeducationscenarioandgender . %DFKHORU V 1

IHPDOH (85

PDOH (85

.HUQHOGHQVLW\ ([SHFWHGQHWPRQWKO\HDUQLQJVLQ(85

0DVWHU

V 1

IHPDOH (85

PDOH (85

([SHFWHGQHWPRQWKO\HDUQLQJVLQ(85

3RROHG

1

IHPDOH (85

PDOH (85

.HUQHOGHQVLW\ )HPDOH 0DOH Notes:Thisfigureshow sthecross-sec tionaldistributi onofexpecte daveragewageswithdi!erentedu- cationaldegreesforme n(blue)andwomen(red ).Obse rvationsabo vethehighestandbelow thelo west percentileoftherespectivedistr ib utionsareex cluded.Forillustrationpurposeswagee xpectationsex- ceeding8,000EURpermontharen otdepi cted.Source:Berliner-Studi enberechtigten-Panelwave3. over15%.Severa lfurtherobse rvationsarenoteworth y:First,averagee xpectedwages increasewiththeleveloft heeducations cenario, indicatingthatstudent sarea wareof themoneta ryreturnstohigherlevelso feducation. 10

Second,thehigherthee ducational

degree,themoredispersed thedistribut ion.Thisshows increasingheterogeneityinex- pectedaveragewages withincreasinglevelsofeducatio n. 11

Third,bothmale andfemale

wagedistributionsa reright-skewed,justas isactualea rningsdistributions. Finally,inalleduca tion scenario sthedistributionofmenisshift edtotherightand exhibitsathickerr ight tail,implyingthatmenexpe cthigherwagestha nwomenonav - erageandaremorelik elytoex pectexcept ionallyhighwag es.AKolmogo rov-Smirno v testconfirmsthat thedistributionsofcr oss-sect ionalave rageexpectedwages di!ersig- nificantlybygenderineac heducatio nscenario.Mor edet aileddescriptives tatisticscan befoundin AppendixTableC.1. 10 InAppen dixSectionCwecompar etheexpectations toactualearn ingsbygender.Interestingly,both gendersoverestimatet hereturnstohigherdegrees,butfemale soveresti matethemost. 11 Thisfindingi sinlinewiththefindi ngsonact uallaborm arketdata(KoerselmanandUusitalo,2014). 7

2.2Samplecharacteri stics

Thedatainc ludedetailedinfo rmationo ncharacteristicst hatarepotentiallyrelated towage expectations. Weoutlinethosevariablesbelowandreportmeansby genderin

Table1.

(I)Baselinecharacteristicscomprisesocio-demogra phicfactorssuchasmigration background,parentaleducationa lbackground,andattendedhighsc hooltype(academic highschoo l,integratedhighsc hool,andvocationalhighschool).Menand womendo notdi!ersignifica ntlyinthesecharacteristics ,wit htheexceptiono fattendedhighschool type.Wealsorepo rtwhethe rstudentsa ttendedaninformationinterventionscho olora financialintervention schooloftheBe stUpproj ect. (II)Cognitiveabilitiescomprisestudents'final highschoolGPAaswellastes t scoresonaverbalandfig uralcog nition test. 12

Wewoulde xpecthigherp erforming

studentstoanticipatethat theirhighe rability(orsignalthereof)isre wardedinthe labor market.Surprisingly,womens corelowerontheverbal cognitiontestby aroundone point.Following(Fortin,2008),weuse standardiz edvalues forallcognitiveabilitiesin thedecomp osition. (III)Intendedcollegemajo raccountsforthewell-document edwagedi!erences betweencollegemajorsaswe llasdi!erencesincollegemajorc hoice bygender.The intendedcollegemajors houldproxythetypeof occupationand/orindus tryindividuals aspiretoworkin andshouldt husreflectdi!erencesinwageexpecta tions duetosorting (Montmarquetteetal.,2002,Arcidiaconoetal.,2020).Bas edontheclassification ofthe GermanStatisticalO"ce(Destatis,2012),thedi!erentmajorsaregr oupedintotenfields ofstudy ,aslistedinTable1. 13 Womenaresignifican tlymorelikelyt oexpressintentions toenroll inmedicalstudiesandt eac hingandarelesslike lytoreportinte ntionsf ora majorinSTEM. (IV)Careermotivescapturetheimportance ofdi!erentjobattributes thatstudents assigntotheirfuturej obchoice .Similartothein tendedcollegemajo r,wageex pectations arelikelyto bea!ectedbythecaree rplansthat students hold,whichinturnarelikelyquotesdbs_dbs15.pdfusesText_21

[PDF] checkpoint charlie

[PDF] peter fechter

[PDF] ich bin ein berliner

[PDF] rda

[PDF] description de l'egypte livre

[PDF] description de l'egypte bibliothèque de l'image

[PDF] description de l'egypte napoleon

[PDF] description de l'egypte numérisée

[PDF] description de l'egypte panckoucke

[PDF] bonaparte egypte

[PDF] interpretation du port de bordeaux

[PDF] vers quel océan mène la garonne

[PDF] vue du port de bordeaux commanditaire

[PDF] qui a fait connaitre la chanson de craonne au public

[PDF] analyse de la première de couverture de candide