What men earn, and women do not : women’s struggles on the Labour market

and Women do not

Women’s Struggles on the Labour market

Rafaël Renaldo van Hout

10280979

rafaelvanhout@me.com

MSc Sociology: Comparative

Organisation & Labour studies

Prof. dr. B.G.M. Völker

Dr. S.M. Steinmetz

Master Thesis

!

The! gender! wage! gap! is! prevalent! in! many! countries! in! the! world.! The! Netherlands!is!no!exception!to!this.!Using!OECD!data!of!2012!this!work!tries! to!explain!the!Dutch!hourly!wage!gap!of!men!and!women!working!in!the!same! occupations.!Three!areas!will!be!focused!on!with!regards!to!hourly!wages:!(I)! the! soHcalled! motherhood! wage! penalty! (i.e.! the! effect! of! being! a! mother! on! hourly! wages),! (II)! the! effect! of! numeracy,! literacy! and! problem! solving! individual! skills! and! how! these! are! valued! differently! between! men! and! women,!and!(III)!the!feminisation!(i.e.!the!proportion!of!women!working!in!an! occupation).! Previous! studies! were! utilised! for! the! theoretical! background! and!these!focused!mostly!on!the!United!States.!The!Dutch!situation,!however,! turns! out! to! be! different! on! several! of! these! three! focus! areas.! The! motherhood! wage! penalty! is! also! present! in! the! Netherlands,! but! from! the! skills!area!only!the!individual!level!of!numeracy!is!in!line!with!what!was!to!be! expected.! The! opposite! effect!was! found!for! the! third!focus! area.!Lastly,! this! thesis!ends!with!a!conclusion!and!recommendations!for!future!research.!

' ' ' First&of&all,&I&want&to&thank&my&supervisor&Professor&Beate&Völker&for&her&help,& enthusiasm,&knowledge,&understanding&and&most&of&all,&patience.& & Another&word&of&thanks&goes&to&Stephanie&Steinmetz.&Not&only&did&she&agree&to&be& my&second&supervisor&for&my&master&thesis,&she&also&taught&me&during&my&bachelor& studies&about&advanced&statistical&methods,&and&also&work&related&gender& inequalities&during&the&course&Social&Stratification,&Life&Courses&&&Gender.& Especially&this&course&inspired&me&to&dedicate&both&my&theses&to&gender&inequality.& & I&also&want&to&thank&everyone&at&the&department&of&sociology&for&broadening&my& views&and&making&me&think&twice&about&society.&Furthermore,&Jonneke,&Suzanne& and&Rosalie&need&to&be&mentioned&here&as&well,&as&they&were&my&foremost& companions&during&my&studies.&Thanks&ladies&for&your&friendship!& & Last&but&not&least,&I&want&to&thank&my&mother&for&how&she&inspires&me&to&always& work&hard&and&follow&your&dreams.&You&are&the&exception&to&many&of&the&gender& inequalities&I&have&studied&during&my&studies.&By&perseverance&you&make&you&career& happen&without&neglecting&your&children.&Much&love&to&you&from&your&son.&

!

'

SECTION!2.3!SKILLS!&!WAGES! 7!

SECTION!2.4!THE!MORE!WOMEN,!THE!MERRIER?! 9!

! CHAPTER!3.!DATA!&!METHODS! 12! SECTION!3.1!DATA! 12! SECTION!3.2!METHODS! 13! SECTION!3.3!VARIABLES! 17! ! CHAPTER!4.!RESULTS! 22! SECTION!4.1!CONSTRUCTION!OF!THE!MODELS! 22! SECTION!4.2!CLOSE!INTERPRETATION!OF!THE!RESULTS! 22! ! CHAPTER!5.!DISCUSSION! 28! SECTION!5.1!FROM!RESULTS!BACK!TO!THEORY! 28! ! CHAPTER!6.!CONCLUSION! 31! SECTION!6.1!CONCLUSION! 31! SECTION!6.2!RECOMMENDATIONS!FOR!FUTURE!STUDIES! 31! SECTION!6.3!REFLECTION! 32! ! REFERENCES! 34! ! APPENDIX!1.!ADDITIONAL!TABLES! 37! APPENDIX!2.!DOFFILE! 40! ! ! '

Chapter(1.(Introduction( !

Why$do$women$earn$less$than$men$while$working$in$the$same$occupation?$

!

Work!related!gender!inequalities,!and!more!specifically!the!gender!wage!gap,!have! been! studied! many! times! by! social! scientists! and! economists.! Even! though! we! currently!live!in!a!modern!society!in!which!women!have!more!or!less!equal!access!to! education,! labour! market! inequalities! between! the! sexes! persist.! For! example,! an! online!article!of!Dutch!newspaper!Trouw!in!which!philosopher!Ingrid!Robeyns!was! interviewed,!mentioned!that!the!work!performed!by!men!is!valued!more!than!that! of!women,!and!that!they!are!thus!getting!higher!rewards!(Kruidhof!2015).!According! to!Robeyns,!“Gender!inequality!exists!on!cultural,!social,!psychological!and!economic! spheres![…]!even!though!it!is!a!form!of!inequality![and]!you!want!people!to!take!note! of!this!and!discuss!it!in!society”!(ibid.).!Using!Eurostat!data!the!article!further!shows! that!the!Netherlands!in!the!regard!of!the!gender!wage!gap!is!mediocre!at!best.!It!is! definitely! not! in! the! worst! performing! group,! although! it! is! not! best! performing! group!either.!Even!so,!a!16!per!cent!difference!in!hourly!wages!between!men!and! women!is!substantial.!An!interesting!point!raised!by!the!article!is!about!the!different! jobs!and!occupations!men!and!women!have,!which!in!turn!could!affect!the!gender! wage!gap!as!some!occupations!simply!have!higher!wages!attached!to!them.! ! At!the!end!of!my!bachelor!studies,!I!too!decided!to!focus!on!the!gender!wage!gap! phenomenon!in!order!to!find!out!why!and!how!exactly!women!tend!to!be!in!worse! off! on! the! Dutch! labour! market! when! compared! to! men.! I! tried! to! find! an! explanation!by!looking!at!differences!between!the!jobs!that!people!have!and!their! individual! characteristics.! The! basic! premise! was! that! men! and! women! have! different! skills! and! that! the! skills! of! men! are! disproportionally! favoured! when! compared! to! that! of! women.! Skills! were! operationalised! using! (the! frequency! of! certain)! work! tasks,! implying! that! doing! something! often! also! means! that! one! is! skilled!at!it.!The!results!were!mostly!unsatisfactory!in!that!no!significant!effects!were! found!whatsoever!in!the!regard!of!skills.!What!did!give!results!was!the!inclusion!of! the!proportion!of!women!working!in!an!occupation.!Another!point!of!interest!was!

the! effect! of! individual! numeracy! skill! level! because! it! stayed! fairly! constant! in! all! models.!Even!though!I!only!added!this!as!a!control!effect,!it!seemed!that!becoming! smarter! (i.e.! more! skilled)! has! a! positive! effect! on! hourly! wages.! Especially! this! particularity!is!the!thing!I!want!to!explore!more!in!depth,!and!it!is!also!this!aspect!of! my!research!in!which!much!more!can!be!explored!in!order!to!solve!the!puzzle!of!the! Dutch!gender!wage!gap.!Additionally,!for!this!year!I!also!wanted!to!look!at!the!effect! of!motherhood!on!earnings.!Becoming!a!mother!usually!means!that!a!woman!leaves! the!labour!market!either!momentarily!or!for!a!long!period!of!time.!It!is!interesting!to! find!whether!this!actually!affects!the!hourly!wages!of!women!as!well.! ! Thus,!the!research!question!is:! ! Why$do$women$earn$less$than$men$even$though$they$work$in$the$same$occupation,$ and$what$are$the$influences$of$skills$and$motherhood?$ $ Just!as!last!year!the!PIAAC!survey!of!2012!for!the!Netherlands!was!used.!The!next! chapter!will!start!with!providing!a!theoretical!framework!including!the!hypotheses! that! will! be! tested.! Chapter! 3! will! continue! with! explaining! the! data! and! methods! utilised.!A!close!interpretation!of!the!results!will!be!given!in!chapter!4,!followed!by! the!discussion!chapter.!The!thesis!will!end!with!a!conclusion!and!recommendations! in!chapter!6.!

Chapter(2.(Theoretical(Framework( !

Section(2.1(A(short(introduction(to(theory(

!

That!women!earn!less!than!men!is!an!established!fact.!As!stated!before,!the!situation! in! the! Netherlands! is! no! exception! to! this.! However,! why! this! happens! can! be! considered! odd! because! of! several! reasons.! Firstly,! labour! market! attachment! is! quite! high! in! the! Netherlands.! According! to! Duyvendak! &! Stavenuiter! (2004),! employment!rate!of!women!was!already!65.2%!in!2002.!Moreover,!the!majority!of! higher!education!students!as!well!as!graduates!are!actually!women!at!the!moment! (CBS!2007).!Logically,!it!were!to!be!expected!that!work!related!gender!inequalities,! including!the!gender!wage!gap,!would!be!lower!when!women!would!become!more! active!on!the!labour!market.!On!the!contrary,!the!reverse!seems!to!be!true.! !

This! chapter! will! focus! on! providing! theoretical! explanations! for! the! gender! wage! gap!using!previous!studies.!The!studies!used!in!this!chapter!mostly!focus!on!data!on! the!United!States!and/or!are!a!bit!dated.!Consequently,!this!thesis!tests!whether!the! results!can!be!found!in!the!Netherlands!as!well.!One!important!distinction!should!be! made!beforehand.!The!gender!wage!gap!can!either!be!defined!as!the!difference!in! average!wages!between!men!and!women,!but!a!perspective!like!that!is!flawed!due! to!the!fact!that!this!difference!can!mostly!be!attributed!to!men!and!women!working! in! different! occupations.! If! one! sex! (most! probably! men)! is! more! represented! in! higher! rewarding! occupations! then! it! obviously! follows! that! a! large! share! of! the! gender! wage! gap! simply! exists! because! of! this.! Consequently,! this! thesis! is! more! concerned! with! gender! wage! differences! between! men! and! women! in! the! same! occupation!rather!than!those!between!men!and!women!in!general.!

!

This!thesis!focuses!on!three!possible!reasons!of!why!women!tend!to!earn!less!than! men! in! the! same! occupation.! The! first! explanation! to! be! explored! is! the! effect! of! having! children! and! especially! how! it! differs! between! men! and! women.! The! importance!here!lies!in!how!the!Dutch!welfare!state!is!designed,!which!cultural!ideas! lie!at!the!basis!of!this!and!how!these!relate!to!cultural!perceptions!of!parenthood!

(and! most! importantly! motherhood).! These! cultural! perceptions! in! turn! determine! the! expectations! employers! have! of! their! employees! as! well! as! how! individuals! themselves!view!parenthood.!Secondly,!the!importance!of!skills!and!how!these!are! valued!differently!between!men!and!women!will!be!reviewed.!Especially!researchers! from!the!human!capital!theory!side!have!strong!ideas!on!the!relation!between!skills,! productivity!and!wages.!However,!other!researchers!have!shown!that!skills!are!not! necessarily! valued! equally! between! men! and! women,! meaning! that! the! human! capital! theory! needs! to! be! revised.! Thirdly,! the! proportion! of! women! in! an! occupation! can! also! be! seen! as! an! important! determinant! of! the! wages! of! said! occupation.! In! principle,! this! can! be! important! in! two! ways:! (I)! the! average! wages! can! be! influenced! by! an! overrepresentation! of! one! gender,! and! (II)! a! woman! in! a! feminised!occupation!might!be!valued!differently!than!a!men!in!the!same!situation! (this! can! also! differ! in! different! settings,! e.g.! being! a! women! in! a! masculine! occupation).!In!the!three!sections!these!explanations!will!be!carefully!explored.! ! ! Section(2.2(The(curse(of(motherhood( ( It!has!been!shown!by!many!studies!that!having!children!negatively!affects!the!hourly! wages! of! women! (see! e.g.! Glauber! 2012).! This! is! referred! to! as! the! motherhood! wage!penalty.!The!effect!is!not!equal!for!every!child!as!Waldfogel!(1997)!found!that! for!one!child!a!3.8%!wage!loss!is!expected,!and!Budig!&!England!(2001)!found!that! one,!two,!and!three!and!more!children!lead!to!losses!of!respectively!3.2%,!8.9%!and! 12.1%!(Glauber!2012:!118).!! ! Many!explanations!have!been!given!for!this.!On!the!one!hand,!Glauber!(2012)!states! that!the!human!capital!theorists!relate!it!to!the!fact!that!women!have!to!leave!the! labour!market!for!a!period!of!time!when!pregnant,!and!that!this!leads!to!loss!of!work! experience!(see!e.g.!Greenman!2011)!as!well!as!hampering!their!skill!formation!(see! e.g.!Kaufman!&!Uhlenberg!2000).!Other!explanations!revolve!around!discrimination! against!mothers!in!terms!of!“hiring,!promotion,!and!evaluation!decisions”,!that!they!

might!be!less!productive!than!men!or!that!“mothers!choose!lower!wages!in!return! for!desirable!working!conditions”!(ibid.:!118).!! ! The!‘varieties!of!capitalism’!theory!also!provides!an!explanation!for!this.!EstévezeAbe! et!al.!(2001)!state!that!employment!protection!for!women!also!entails!policies!that! help!women!return!to!the!job!(2001:!158).!For!them,!!

“[the]! two! factors! are! (i)! protection! against! dismissal,! such! as! maternity,! parental,!and!family!leave!policies;!and!(ii)!income!maintenance!during!leaves! and! guarantees! of! reinstatement! to! the! same! wage! level! upon! return! to! work”!(ibid.:!159).!!

It! is! because! of! these! seemingly! positive! protections! that! women! will! have! lower! wages!eventually.!Similarly!to!what!the!human!capital!theorists!argue,!when!women! get! “[reinstated]! to! the! original! job! after! the! leave! means! that! women! fall! behind! their!male!cohort!in!skill!formation!and!promotion”,!even!when!still!doing!the!same! job!(ibid.:!158e9).!Secondly,!EstévezeAbe!(2005)!points!out!that!the!most!important! moment! in! one’s! career! to! increasingly! develop! useful! skills! corresponds! with! the! years! women! usually! become! a! mother! (2005:! 190).! Thirdly,! it! is! quite! costly! for! employers!to,!on!the!hand,!pay!for!a!women’s!absence,!and!on!the!other,!pay!for! their! replacements! (EstévezeAbe! et! al.! 2001:! 159).! Because! of! this,! their! relative! costs!also!become!higher!than!that!of!men.!!

!

Most! of! the! studies! mentioned! before! focus! solely! on! the! US.! It! thus! remains! the! question!whether!the!motherhood!wage!penalty!can!be!found!in!the!Netherlands!as! well!and!in!what!degree.!Mandel!&!Semyonov!have!related!the!gender!wage!gap!to,! on! the! one! hand,! family! policies,! and! on! the! other! wage! structures.! According! to! them,! “‘motherefriendly’! [family]! policies! enable! more! women! to! become! economically!active”!and!countries!with!these!policies!are!known!to!have!a!smaller! gender!wage!gap!(2005.:!949).!However,!this!is!not!because!of!those!family!policies! but!actually!despite!of!them!–!the!actual!reason!why!those!countries!have!smaller! gender!wage!gaps!is!because!of!“more!egalitarian!wage!structures”!(ibid.).!Similarly! to! what! others! also! said,! leave! policies! can! lead! to! “long! absence! from! paid! employment! [that]! may! reduce! women’s! earning! capacity! by! lowering! their!

employment!continuity!and!work!experience”!(ibid.:!951).!Especially!if!these!policies! are!aimed!exclusively!on!women,!men!will!enjoy!a!stable!advantage!over!women!in! their! professional! careers.! In! turn,! employing! mothers! for! the! same! wage! is! less! desirable!and!they!thus!have!lower!hourly!wages.!

!

Clearly,!an!advantage!of!Mandel!&!Semyonov!(2005)!over!what!Glauber!(2012)!said! is! that! the! former! is! more! situational.! Mandel! &! Semyonov! looked! at! multiple! countries! at! once,! while! Glauber! and! the! studies! she! quoted! mostly! gave! general! explanations! that! were! subsequently! tested! with! data! on! the! US.! Mandel! &! Semyonov,! however,! did! not! test! specifically! for! being! a! parent! (i.e.! a! mother),! although!this!does!not!mean!that!it!is!irrelevant.!According!to!them,!the!Netherlands! can! be! seen! as! part! of! the! conservative! welfare! states! (ibid.:! 957).! These! welfare! states! generally! have! more! family! policies! than! the! AngloeSaxon! liberal! welfare! states!(e.g.!the!US).!The!archetypical!conservative!welfare!state!uses!social!policy!in! order! to! keep! certain! traditions! and! cultural! values! in! place! (see! e.g.! Espinge Andersen!1990!for!an!extensive!overview!on!welfare!state!regimes).!These!countries! were! traditionally! known! for! their! policies! that! promoted! the! breadwinner! model! (e.g.!Germany).!!

!

Traditionally,! in! the! Netherlands! it! was! the! husband’s! responsibility! to! earn! the! money! on! which! the! family! lives,! and! the! wife’s! responsibility! to! take! care! of! the! family,! but! it! has! since! then! evolved! in! a! oneeandeaehalf! breadwinner! model! (in! which!women!are!culturally!‘allowed’!to!work!whilst!their!main!satisfaction!derives! from! their! ‘duties’! at! home)! (Hofmeister! et! al.! 2006:! 14,! 16).! This! is! achieved! by! implementing! those! family! policies! (e.g.! maternity! leave,! sick! care),! as! women! are! more! able! to! combine! work! and! family! life.! However,! the! family! policies! in! the! Netherlands! have! generally! been! implemented! unequally.! For! example,! maternity! leave! is! much! higher! than! the! time! men! get! off! work! when! their! child! is! born.! According!to!Hofmeister!et!al.,!equal!pay!between!men!and!women!will!probably!not! come! into! existence! as! long! as! family! policies! do! not! equalise! the! differences! between! men! and! women! (Hofmeister! et! al.! 2006:! 16).! Similarly,! women! are! still! also!the!ones!that!are!easily!able!to!take!care!of!the!children!when!needed,!and!men!

are!not.!Dutch!men,!even!though!they!believe!that!their!partners!and!themselves!are! equally! good! at! parenting,! still! face! cultural! dilemmas! (Duyvendak! &! Stavenuiter! 2004:! 17).! In! particular,! “men! have! to! work! up! to! the! image! of! the! hard! working! male”!and!“for![them]!the!concept!of!care!giving!is!problematic,!because!care!lacks! the! symbolic! value! of! paid! labour,! which! is! associated! with! status,! power! and! income”!(ibid.:!21).!Thus,!having!children!means!that!Dutch!mothers!still!have!to!do! most!of!the!caring!tasks!at!home!and!leave!the!labour!market!more!than!men.!Even! if!mothers!do!not!need!to!leave!their!jobs!regularly,!they!are!probably!still!expected! by!employers!to!be!less!engaged!with!their!work!than!their!male!cohorts.! ! In!sum,!the!first!hypothesis!is:!

1. Having( children( will( negatively( impact( the( hourly( wages( of( women( when( compared(to(men.(

'

Although! this! hypothesis! is! more! exploratory! than! explanatory,! it! still! suffices! because!this!thesis!aims!at!measuring!the!wage!differences!in!the!Netherlands.!The! theories!and!explanations!from!above!did!not!specifically!on!the!Netherlands,!and!it! is! thus! primarily! interesting! whether! the! motherhood! wage! penalty! exists! in! the! Netherlands.!Moreover,!the!dataset!utilised!here!cannot!easily!be!used!for!testing!all! explanations!from!the!above.!Further!research!is!needed!for!this.! ! ! Section(2.3(Skills(&(wages( ( Relating!wages!to!something!as!elementary!as!skills!has!been!done!multiple!times.! One!welleknown!theory!to!explain!this!is!the!human!capital!theory!(Becker!1964).!In! principle,! the! human! capital! sees! a! direct! relation! between! skills! and! wages.! Independent!of!what!type!of!skill!is!being!observed,!the!better!one’s!skills!are,!the! more! productive! he! or! she! are,! and! the! higher! the! wages! will! be.! According! to! human! theory! supporters,! this! would! be! true! because! how! better! skilled! and! productive! an! individual,! the! higher! the! demand! for! him! or! her! will! be.! The! consequence!of!this!is!that!the!gender!wage!gap!exists!because!women!merely!have!

less!developed!skills!than!men.!In!this!sense,!it!is!not!far!off!from!what!EstévezeAbe! et! al.! said! about! women! lagging! behind! in! skill! formation! when! returning! from! maternity!leave!(2001:!158e9).!!!

!

However,! the! human! capital! approach! has! been! refuted! more! than! once.! For! example,!Kilbourne!et!al.!created!hypotheses!using!both!neoeclassical!theories!(e.g.! the! human! capital! theory)! and! a! cultural! feminist! theory.! According! to! the! human! capital!theory,!!

“employers!have!to!pay!more!for!skilled!workers!if!gaining!these!skills!was!an! ‘investment’!(i.e.,!involved!a!present!cost!for!a!potential!future!benefit)!that! the!marginal!worker!would!not!make!without!compensation”!(1994:!690).! Kilbourne! et! al.’s! answer! to! the! human! capital! theory! is! the! “theory! of! gendered! valuation”,!its!basic!point!being!that!the!gender!wage!inequality!is!structured!socially! and!not!a!pure!necessity!(1994:!695).!Its!main!explanation!“is!that!cultural!processes! of! valuation! are! gendered;! because! women! are! devalued,! social! roles! (including! occupations)! and! skills! that! are! associated! with! women! are! culturally! devalued! relative! to! those! associated! with! men”! (ibid.).! Their! findings! underline! their! gendered!valuation!hypothesis.!Kilbourne!et!al.!(ibid.)!have!controlled!for!four!types! of! skills! in! their! regression! of! wages,! namely! cognitive,! physical,! authorative! and! nurturant! skills! (1994:! 701).! Cognitive,! physical! and! authorative! had! positive! coefficients!for!men,!while!women!only!had!that!for!the!first!type!of!skill!(which!is! actually!higher!than!that!of!men).!Nurturant!skill,!the!typically!female!skill!added!in! the!model,!lowered!wages!for!both!men!and!women,!and!even!does!so!stronger!for! the!latter!(ibid.).! ! Due!to!the!data!at!hand!(which!will!be!described!in!the!next!chapter)!other!skills!will! be!tested!in!this!thesis.!The!PIAAC!survey!tests!for!three!types!of!skills:!numeracy,! literacy! and! problem! solving.! Of! these,! numeracy! can! be! perceived! as! a! typically! male!type!of!skill!as!young!boys!are!gendered!towards!it!from!a!young!age!onwards! and!young!girls!are!not!(see!e.g.!Spelke!2005).!In!another!study,!Mitra!(2002)!found! that!in!the!US!whiteecollar!women!earn!less!than!men!with!the!same!mathematics! score,!although!the!effect!is!insignificant!for!blueecollar!workers,!and!even!reversed!

for! professionals,! managers! and! highly! skilled! workers! (2002:! 449e50).! In! all! these! instances!(except!that!for!blueecollar!workers),!the!effect!is!always!positive,!meaning! that!being!good!in!mathematics!is!almost!always!expected!to!be!beneficial!for!one’s! wages.!Furthermore,!!

“the! results! [found! by! Mitra]! show! that! (1)! quantitative! skills! are! in! high! demand!in!the!labor!market,!(2)!mathematics!skill!leads!to!significant!wage! premiums!for!women,!and!(3)!the!gender!wage!gap!is!not!significant!among! professional!men!and!women!with!aboveeaverage!mathematics!skills”!(ibid.:! 455).!! ! It!thus!means!that!it!can!be!expected!that!the!total!effect!of!numeracy!on!wages!is!at! least!positive!for!both!men!and!women!although!stronger!for!men.!Because!literacy! is!a!typically!female!skill,!it!is!expected!that!the!total!effect!of!literacy!on!wages!is!at! least! negative! for! both! men! and! women! as! well! as! that! this! negative! effect! is! stronger! for! the! latter! group.! Because! it! is! a! cognitive! skill,! problem! solving! is! expected! to! be! positive! for! both! men! and! women,! although! Kilbourne! et! al.! also! found!that!women!have!higher!returns!than!men!(1994:!701e2).!They!do!not!give!an! explanation!for!this.! ! In!sum,!hypotheses!2,!3!and!4!are! 2. Having(literacy(skills(lowers(the(hourly(wages(of(both(men(and(women,(and( this(effect(is(stronger(for(women.( 3. Having(numeracy(skills(increases(the(hourly(wages(of(both(men(and(women,( although(this(is(less(strong(for(women.( 4. Having(problem(solving(skills(increases(the(hourly(wages(of(both(men(and( women,(and(this(is(effect(is(stronger(for(women.( ! ! Section(2.4(The(more(women,(the(merrier?( ! Another!gendered!valuation!effect!to!be!tested!is!the!feminisation!of!an!occupation.! The!basic!idea!is!that!working!in!occupations!with!a!high!proportion!of!females,!has!a!

negative!effect!on!wages!in!general!and!on!that!of!women!specifically.!!According!to! Kilbourne!et!al.,!!

“such!devaluation!may!be!the!cognitive!error!of!not!seeing!how!much!female! occupations!contribute!to!the!‘bottom!line’!of!profit!or!other!organizational! goals.! Or! women’s! jobs! may! be! devalued! because! they! are! erroneously! believed!to!require!less!effort!than!men’s!jobs,!because!typically!female!types! of!effort!are!culturally!invisible.!Once!interoccupational!wage!differences!are! in!place,!institutional!inertia!keeps!genderebiased!interjob!wage!difference!in! force”!(1994:!695).!

Cohen! &! Huffman! (2003)! add! to! this! that! men! can! be! seen! as! scarce! in! female! occupations!thus!enabling!them!to!better!career!possibilities!than!women!in!those! occupations! (2003:! 445e6).! ! This! means! that! men,! although! they! are! expected! to! have! lower! wages! than! men! working! in! male! occupations,! still! have! higher! wages! than!women!in!female!occupations.!In!other!words,!!

“[women]! in! female! jobs! thus! pay! two! penalties:! Not! only! is! the! average! wage! lower! than! that! for! comparable! maleedominated! jobs,! they! also! earn! less!relative!to!men!in!the!same!jobs”!(ibid.:!457).!

This!is!in!line!with!the!models!presented!by!Kilbourne!et!al.!(1994).! !

Generally!speaking,!it!can!thus!be!expected!that!the!same!effect!will!be!found!in!the! Netherlands.!However,!the!data!used!by!Cohen!&!Huffman!(2003)!and!Kilbourne!et! al.! (1994)! only! applies! on! the! US! and! is! quite! some! time! in! the! past! already.! Furthermore,!Cohen!&!Huffman!state!that!wage!inequality!happens!at!the!job!level! in!specific!states!and!not!at!the!occupational!level!meaning!that!the!data!itself!can! be! distorted! and! local! effects! excluded! from! the! analysis! (2003:! 445e6).! In! the! Netherlands,!however,!collective!bargaining!happens!on!a!large!scale,!and!this!leads! a!centralised!wage!system!for!occupations.!Because!both!the!place,!time!and!wage! setting!situation!is!different!there,!it!is!quite!clearly!possible!that!women!do!not!face! the! “two! penalties”! (ibid.:! 457).! In! that! sense,! the! theory! of! tokenism! might! be! a! better!explanation!(Kanter!1977).!In!principle,!that!theory!states!that!!

“increased! concentrations! of! women! will! decrease! gender! inequality! …! [because]! men’s! perception! of! women! is! conditional! on! an! organization’s!

gender!composition!–!when!women’s!representation!increases,!women!face! less! performance! pressure,! have! more! power! than! women! in! malee dominated!work!settings,!and!therefore!suffer!less!discriminatory!treatment”! (Cohen!&!Huffman!2003:!445).! However,!in!principle!this!thesis!will!test!what!Cohen!&!Huffman!found,!namely!that! men!earn!more!than!women!in!feminine!jobs.! ! Thus,!hypotheses!5!and!6!are! 5. The(average(hourly(wages(are(lower(in(occupations(with(a(high(proportion( of(women.(

6. The( hourly( wage( gap( between( men( and( women( becomes( larger( in( occupations(with(a(high(proportion(of(women.(

Chapter(3.(Data(&(Methods( ! Section(3.1(Data( ! For!the!research!conducted!in!this!thesis!the!same!data!was!used!as!last!year.!The! PIAAC!survey!(Programme!for!the!International!Assessment!of!Adult!Competencies),! or! Survey! of! Adult! Skills! for! short,! was! set! up! by! the! Organisation! of! Economic! Cooperation! and! Development! (OECD)! in! order! to! help! member! states! assess! the! competencies!of!their!inhabitants!(i.e.!their!skills!and!their!skill!use!at!work!and!at! home)!(OECD!2015a).!The!data!is!collected!in!more!than!thirty!countries!in!a!similar! way.!In!turn,!the!results!can!be!used!for!creating!effective,!national!policy,!as!well!as! comparing! the! performances! of! different! countries.! In! this! sense,! it! connects! seamlessly! to! the! OECD’s! mission! “to! promote! policies! that! will! improve! the! economic!and!social!wellLbeing!of!people!around!the!world”!(OECD!2015b).!As!the! PISA!survey!focuses!on!the!performances!of!children,!the!newly!established!PIAAC! dataset! can! be! seen! as! its! adult! version.! Both! PIAAC! and! PISA! are! intended! to! be! longitudinal! surveys,! meaning! that! the! changes! between! survey! rounds! can! be! observed.!In!contrary!to!PISA,!PIAAC!has!yet!to!have!its!second!round!finished.! !

This!dataset!has!several!advantages!over!other!conventional!datasets!like!Eurostat.! Fore!mostly,!the!PIAAC!survey!contains!many!questions!about!how!their!work!can!be! organised.! For! example,! some! questions! relate! to! how! many! times! per! week! an! individual! has! to! use! calculator! or! has! to! solve! difficult! mathematical! problems.! More!importantly!for!this!thesis!is!that!the!PIAAC!dataset!contains!information!on! actual! skill! levels.! Respondents! were! tested! using! computers! (or! with! pencil! and! paper!if!he!or! she! was! not!able! to! do!it! with! a!computer)!in!order!to!assess! their! numeracy,!literacy!and!problem!solving!capabilities.!To!have!accurate!data!on!this,! respondents!only!made!a!fraction!of!the!tests.!The!actual!test!was!extensive!in!such! a!way!that!it!would!take!hours!if!not!days!to!complete.!Plausible!values!were!thus! used,! meaning! that! it! was! calculated! what! a! respondent! would! score! on! test! questions!he!or!she!did!not!answer.!Naturally,!the!PIAAC!dataset!also!includes!many! questions!for!obtaining!background!information.!

!

Although!many!countries!were!available!for!analysis,!only!the!data!for!the!first!round! of! the! Netherlands! (2012! is! used.! This! is! because! of! several! reasons:! (I)! personal! interest!is!highest!in!the!Netherlands,!(II)!hourly!wages!need!to!be!corrected!before! they! can! be! compared! crossLnationally! (different! economic! conventions! exist! between!countries,!and!(III)!using!more!than!a!single!country!adds!another!layer!of! difficulty!to!the!model!(this!will!be!explained!later!in!this!chapter).!The!Dutch!sample! was! provided! by! Statistics! Netherlands! (Dutch:! CBS! or! ‘Centraal! Bureau! voor! de! Statistiek),! a! government! agency! that! collects! statistical! data! for! different! kinds! of! purposes.! The! Statistics! Netherlands! is! one! of! the! agencies! that! besides! having! access! to! the! Dutch! Taxes! &! Revenues! administration! (Dutch:! Belastingdienst)! for! research!purposes,!can!also!look!into!the!municipal!housing!administration!(Dutch:! Gemeentelijke! Basis! Administratie).! This! is! a! nationLwide! administrative! system! in! which! municipalities! record! the! addresses! of! their! inhabitants.! Of! the! 10,000! originally!selected!for!the!sample!approximately!51%!agreed!to!participate,!and!the! sample! was! deemed! representative! for! the! population.! All! respondents! had! to! be! between!the!ages!of!16!and!65!and!the!interviews!were!done!at!their!homes.!! ! ! Section(3.2(Methods( ( The!chosen!statistical!method!is!linear!multilevel!modelling!and!the!programme!used! is! Stata.! Why! linear! regression! is! chosen! will! be! obvious,! however,! the! multilevel! modelling! deserves! an! explanation.! The! most! standard! form! of! regression! with! an! interval!variable!like!hourly!wages!would!be!ordinary!least!squares!(OLS)!regression.! However,! OLS! regression! rests! on! five! important! assumptions:! (I)! linearity! of! the! dependent! variables! and! independent! variables,! (II)! mean! independence! (i.e.! the! mean! of! the! error! terms! does! not! depend! on! the! independent! variables),! (III)! homoscedasticity! (i.e.! same! variance! over! all! values! of! the! independent! variables,! (IV)! uncorrelated! residuals! (i.e.! the! respondents! in! the! dataset! are! chosen! independently!from!one!another),!and!(V)!error!normality!(Allison!1999:!123L31).!Of! these!five,!the!fourth!is!problematic!when!looking!at!hourly!wages!as!wages!are!set!

in!occupations!and!jobs.!For!instance,!the!difference!in!wages!of!schoolteachers!will! be! relatively! small! (especially! because! that! profession! has! a! collective! bargaining! agreement!in!place),!and!the!same!can!be!expected!for!cleaners,!or!retail!employees.! The!violation!of!this!assumption!leads!to!lower!standard!errors,!meaning!that!effects! will!be!generalised!too!quickly.!In!other!words,!“there!will!be!a!tendency!to!conclude! that! relationships! exist! [between! the! dependent! and! independent! variables]! when! they! really! don’t! (ibid.:! 129).! This! is! also! referred! to! as! a! ‘type! I! error’! (Agresti! &! Finlay!2009:!160).!Other!examples!that!probably!violate!the!uncorrelated!residuals! assumption!include!measuring!school!performance!of!students!(which!are!obviously! nested!in!schools!and!classrooms)!(ibid.:!523L4),!and!interviewing!couples!in!surveys! (Allison!1999:!128).!It!has!been!shown!multiple!times!already!that!the!performance! of!a!student!is!highly!dependent!of!his!or!her!class!and!the!teacher’s!effectiveness! and!how!well!the!school!performs.!Couples,!on!the!other!hand,!tend!to!have!highly! corresponding!ideas!and!will!thus!probably!give!the!same!answers!when!asking!for! their! opinions.! Because! students! and! workers! are! nested! in! respectively! the! overarching!abstract!entities!schools!and!occupations,!some!statisticians!prefer!the! term!hierarchical!models!(e.g.!Allison!1999:!181).!!

!

A! standard! linear! relation! would! be! described! in! an! equation! as! the! following:!! !! = ! + !!! + !!;! in!which!the!following!goes:! !_!!is!the!value!of!the!dependent!variable!for!individual!!!(here!hourly!wages);! !!is!the!constant!or!!Lintercept;! !!is!the!coefficient!for!independent!variable!!;! !_!!is!the!value!for!!!of!individual!!! and,!!!!is!the!residual!for!individual!!.!

Besides! accounting! and! correcting! for! the! fact! that! individuals! work! in! different! occupations,! multilevel! models! enable! researchers! to! add! random! effects! to! the! equation.!For!example,!the!intercept!(i.e.!!)!might!be!different!for!every!occupation,! meaning!that!the!predicted!regression!line!of!an!occupation!is!higher!or!lower!than! others.! In! addition! to! the! random! intercepts,! random! effects! can! also! be! created,! meaning! that! the! effect! of! a! variable! (i.e.! the! ! !coefficient)! differs! between!

occupations.! In! principle,! multilevel! models! provide! researchers! with! the! grand! mean!intercepts!and/or!grand!mean!effects.!These!are!the!general!value!of!an!effect! between!all!occupations.!The!random!effect!is!calculated!in!one!value!as!well,!not!in! a! specific! value! for! every! occupation.! This! is! called! the! standard! deviation! of! a! random! effect! with! regards! to! its! mean.! More! interestingly! is! the! variance! of! the! random! effect.! Dividing! the! variance! of! the! random! parameters! with! the! total! variance! (which! is! the! variance! of! the! individual! errors! and! the! variance! of! the! random!effects)!gives!an!insight!about!how!much!of!the!total!variance!is!explained! through! the! fact! that! individuals! are! nested! in! a! higher! level.! This! is! called! the! intraclass!correlation!coefficient!or!‘ICC’!for!short.!Specifically!for!this!thesis,!the!ICC! refers!to!the!how!much!difference!in!incomes!between!individuals!can!be!explained! simply!because!of!the!fact!they!work!in!different!occupations.!It!is!expected!that!the! ICC!becomes!smaller!when!more!level!1!variables!(effects!that!only!differ!between! occupations,! not! within)! are! added.! Consequently,! an! ICC! of! 0! means! that! using! multilevel!models!is!near!pointless,!and!OLS!regression!will!probably!give!the!same! or! very! similar! results.! An! ICC! of! 1,! on! the! other! hand,! means! that! individual! characteristics!do!not!account!for!different!hourly!wages,!and!that!this!can!only!be! explained!because!of!the!fact!that!individuals!work!in!different!occupations.!

!

As!mentioned!before,!the!levels!of!a!multilevel!model!imply!that!individuals!of!one! level!are!nested!in!groups!at!a!higher!level!(e.g.!students!in!schools,!individuals!in! occupations).! As! already! mentioned! partly! before,! both! variables! that! provide! personal! information! (e.g.! gender,! education)! as! well! as! variables! that! only! differ! between!occupations!can!be!added!(e.g.!average!share!of!females!in!an!occupation).! Just!as!last!year,!the!models!in!this!thesis!make!use!of!random!intercept!models.!This! is!because!that!even!if!all!variables!would!be!kept!constant,!it!is!believed!that!wages! between! occupations! will! always! differ! to! some! extent.! In! other! words,! some! occupations!will!simply!always!pay!more!than!others!(e.g.!executives!will!always!earn! more! than! their! most! basic! employees).! However,! random! effects! for! the! independent!variables!are!more!difficult!to!explain!and!to!understand.!For!example,! a!random!effect!for!the!average!proportion!of!females!in!an!occupation!would!mean! that! this! effect! depends! on! the! occupation! (because! that! determines! the!!!value,!

which!is!the!same!for!all!individuals!in!one!occupation),!the!grand!mean!effect!(the! average!!!coefficient! for! all! individuals! in! all! occupations),! and! fact! that! the! effect! can! be! random! between! occupations! (i.e.! even! though! two! occupations! might! equally!have!the!same!average!proportion!of!females!working!in!them,!it!could!still! be! that! the! effect! differs! between! occupations.! After! testing! random! effects! for! gender!and!the!three!individual!skills!(numeracy,!literacy!and!problem!solving)!it!was! decided!not!to!add!random!effects!for!the!independent!variables.!This!is!because!the! random!effect!of!gender!was!not!deemed!positive!in!any!of!the!models!that!included! that! variable,! and! even! though! the! random! effects! of! the! skills! actually! were! positive,!their!standard!deviations!were!so!incredibly!small!that!the!actual!random! effects! were! practically! invisible.! This! is! probably! because! of! multicollinearity! between!some!of!the!independent!variables!(see!e.g.!tables!4!and!5!in!appendix!1).! Thus,!only!random!intercepts!are!used!in!the!models.!The!below!is!a!representation! of!the!mathematical!equation!of!the!models!in!this!thesis!and!that!of!last!of!year:! “!"#"$!0! !"#!$!#%&'( :!!!" = !!!+ !!!!"+ !!"! !"#"$!1! !""#$%&'!() : !!! = !! + !!!!+ !!!! !"##!!"#$%&'(:!!_!" = !_!+ !_!!_!"+ !_!!_! + !_!!+ !_!"! where!goes:!! o !!"!is!the!estimated!hourly!earnings!of!individual!!!in!occupation!!;!! o !!!!is!the!constant!in!occupation!!;! o !!!is!the!coefficient!for!individual!level!variable!!!";!

o !!!is! the! average! value! of! the! constant! for! all! occupations! (i.e.! the! grand!mean!intercept);!!

o !!!is!the!coefficient!of!occupational!level!variable!!!;!

o !!!!is!the!error!of!occupation!!!from!the!grand!mean!intercept;! o !!"!is!the!error!of!individual!!!in!occupation!!”!(Van!Hout!2014:!14).! The!models!are!inserted!in!a!stepwise!fashion.!This!means!that!the!first!model!is!null! model! without! any! independent! variables,! followed! by! a! model! that! only! includes! gender.!The!next!model!includes!the!control!variables,!and!all!subsequent!ones!are! for!testing!the!hypotheses.!The!models!and!the!results!will!be!discussed!extensively! in!the!next!chapter.!The!doLfile!can!be!found!in!appendix!2.!

Section(3.3(Variables(

!

As!already!discussed!earlier!in!this!chapter,!the!PIAAC!dataset!for!the!Netherlands!of! 2012!was!used.!The!first!thing!to!do!in!terms!of!data!selection!was!the!deletion!of! the!unemployed!and!those!that!do!not!want!to!work.!Even!though!it!can!be!agreed! upon! that! those! can! equally! be! interesting! in! terms! of! gender! inequality,! those! cannot! be! used! in! this! analysis.! Of! the! original! 5,170! respondents! 1,227! fell! off! initially!by!only!looking!at!employment!status.!The!descriptive!statistics!can!be!found! in!table!1!below.! ! Table!1.!Descriptive!statistics.! ! Mean! S.D.! Range! Highest!level!of!education! ! ! ! - Primary! 0.046! 0.210! 0!–!1! - Lower!secondary! 0.243! 0.429! 0!–!1! - Higher!secondary!and!vocational! 0.357! 0.479! 0!–!1! - Tertiary!professional! 0.239! 0.426! 0!–!1! - Tertiary!academic! 0.114! 0.318! 0!–!1! Experience! 18.791! 11.749! 0!–!47! Squared!experience! 491.112! 503.895! 0!–!2209! ISEI!score! 45.807! 15.835! 10!–!89! ! ! ! ! Gender!(ref.:!male)! 0.503! 0.500! 0!–!1! Having!children!(ref.:!no!children)! 0.605! 0.489! 0!–!1! Literacy!score! 294.832! 40.966! 131.708!–!440.203! Numeracy!score! 291.241! 42.825! 131.641!–!424.254! Problem!solving!score! 291.633! 38.230! 150.819!–!447.153! Average!female!proportion!in!occupation! ! 0.497! 0.253! 0!–!0.946! Hourly!earnings! 17.220! 8.674! 3.07!–!80.23! ln!hourly!earnings! 2.730! 0.492! 1.122!–!4.386! ! Source.!–!!PIAAC!2012!NL,!own!calculations.!N!=!2,952.! ! !

The!dependent!variable!was!hourly!wages.!The!choice!of!hourly!wages!over!monthly! or! annual! earnings! is! because! of! two! reasons:! (I)! many! Dutch! women! work! partL time,!impairing!their!monthly!and!annual!earnings!when!compared!to!those!of!men.! Several!other!steps!had!to!be!taken!before!the!variable!was!ready!for!use.!Firstly,!a! decision!had!to!be!made!whether!the!hourly!wage!variable!with!or!without!bonuses! included!would!be!used.!After!testing!it!was!found!that!the!differences!were!small,! and! therefore! the! hourly! wage! variable! without! bonuses.! Secondly,! to! avoid! potential! outliers! the! range! was! set! at! 3.07€! to! 81€.! This! corresponds! with! what! would!be!the!minimum!wage!for!16!year!olds!and!a!very!decent!executive!pay.!To! not!let!the!higher!values!of!hourly!wages!affect!the!estimates,!the!natural!logarithm! of!hourly!wages!was!used.!

!

The! variable! with! all! the! occupations! in! it! was! created! using! the! ISCOL08! variable! already! present! in! the! data.! ISCOL08! stands! for! the! International! Standard! Classification! of! Occupations! of! 2008,! and! the! International! Labour! Organization! (ILO)! invented! it! in! order! to! have! a! standardised! framework! for! measuring! the! respondents’!occupations!(ILO!2015a).!The!most!recent!classification!is!that!of!2008.! It!works!with!a!digit!system,!and!practically!this!means!that!occupations!are!divided! into!units!that!belong!to!a!minor!group.!This!minor!group!is!part!of!subLmajor!group! that!in!turn!belongs!to!one!of!ten!major!groups!(ILO!2015b).!So!if!one!respondent! has! the! value! 1234,! he! or! she! works! in! occupation! 1234! that! is! part! of! the! minor! group!123,!subLmajor!group!12!and!major!group!1.!Consequently,!the!most!precise! occupational!determination!would!be!the!one!with!four!numbers,!which!is!called!the! 4Ldigit! classification.! Conversely,! the! major! group! is! the! least! precise! one! and! is! referred!to!as!the!1Ldigit!classification.!Logically,!the!4Ldigit!classification!would!be! used! but! there! is! a! catch.! Multilevel! modelling! assumes! that! the! observations! per! level!1!group!variable!is!at!least!around!20.!Only!the!2Ldigit!and!1Ldigit!classifications! adhere!to!this,!meaning!that!the!2Ldigit!was!chosen!for!this!thesis.!Some!of!the!two! digit!values!had!not!enough!observations!so!they!were!recoded!and!added!to!similar! occupations.! This! was! based! on! what! was! done! last! year! and! can! be! found! in! the! table!below.!

Table!2.!ISCOL08!twoLdigit!transformation.!

! Occupational!group! N[old]! N[new]!

02.! NonLcommissioned!armed!forces!officers! 1! Excluded! 03.! Armed!forces!occupations,!other!ranks! 8! Excluded! 11.! Chief!executives,!senior!officials!and!legislators! 117! 117! 12.! Administrative!and!commercial!managers! 114! 114! 13.! Production!and!specialised!services!managers! 228! 228! 14.! Hospitality,!retail!and!other!service!managers! 63! 63! 21.! Science!and!engineering!professionals! 89! 89! 22.!! Health!professionals! 135! 135! 23.! Teaching!professionals! 236! 236! 24.! Business!and!administration!professionals! 179! 179! 25.! Information!and!communications!technology!professionals! 104! 104! 26.! Legal,!social!and!cultural!professionals! 146! 146! 31.! Science!and!engineering!associate!professionals! 115! 115! 32.! Health!associate!professionals! 107! 107! 33.! Business!and!administration!associate!professionals! 290! 306! 34.! Legal,!social,!cultural!and!related!associate!professionals! 128! 128! 35.! Information!and!communications!technicians! 16! Merged!with!33.! 41.! General!and!keyboard!clerks! 118! 118! 42.! Customer!and!services!clerks! 92! 92! 43.! Numerical!and!material!recording!clerks! 181! 181! 44.! Other!clerical!support!workers! 52! 52! 51.! Personal!service!workers! 135! 135! 52.! Sales!workers! 242! 242! 53.! Personal!care!workers! 204! 204! 54.! Protective!service!workers! 63! 63! 61.! MarketLoriented!skilled!agricultural!workers! 47! 52! 62.! MarketLoriented!skilled!forestry!and!fishery!workers! 4! Merged!with!61.! 63.! Subsistence!farmers,!fishers,!hunters!and!gatherers! 1! Merged!with!61.! 71.! Building!and!related!trades!workers! 92! 92! 72.! Metal,!machinery!and!related!trades!workers! 68! 68! 73.! Handicraft!and!printing!workers! 9! 62! 74.! Electrical!and!electronic!trades!workers! 42! 42! 75.! Food!processing,!wood!working,!garment!and!other!craft!and!related!trades!workers! 53! Merged!with!73.! 81.! Stationary!plant!and!machine!operators! 24! 33! 82.! Assemblers! 9! Merged!with!81.! 83.! Drivers!and!mobile!plant!operators! 86! 86! 91.! Cleaners!and!helpers! 106! 133! 92.! Agricultural,!forestry!and!fishery!labours! 25! 75! 93.! Labourers!in!mining,!construction,!manufacturing!and!transport! 93! 93! 94.! Food!preparation!assistants! 27! Merged!with!91.! 95.! Street!and!related!sales!and!service!workers! 1! Merged!with!92.! 96.! Refuse!workers! 49! Merged!with!92.! Source.!–!PIAAC!2012!NL,!own!calculations!but!from!Van!Hout!(2014).!Note!~!PreLselection!on!basis!of!employment! status! and! hourly! earnings! happened! before! the! ISCO! variable! transformations.! Further! selection! happened! afterwards.!Actual!amount!at!N[new]!used!at!analysis!is!slightly!smaller!with!N[61]=17!as!the!smallest.!

The!control!variables!used!were!highest!level!of!education,!years!of!work!experience! and!the!prestige!level!of!the!job!a!respondent!has.!The!original!variable!of!education! included! 16! different! categories! ranging! from! no! degree! at! all! to! having! a! PhD! degree.!I!recoded!these!into!five!categories:!(I)!primary!education!or!lower,!(II)!lower! secondary! education! (Dutch:! VMBO),! (III)! higher! secondary! (Havo! or! VWO)! or! vocational!education,!(IV)!university!of!applied!sciences!(Dutch:!HBO!Bachelor),!and! (V)! university! (WO! Bachelor! and! higher).! Corresponding! foreign! degrees! were! recoded! to! these! as! well.! The! variable! of! experience! ranges! from! 0! to! 47! and! its! square!was!added!as!well.!This!was!done!because!the!effect!of!having!30!or!40!years! of!experience!changes!much!less!than!that!of!0!and!10!years!of!experience.!Adding! both! the! normal! and! the! squared! variable! of! work! experience! corrects! this.! An! additional! control! variable! was! the! International! SocioLEconomic! Index! of! Occupational!Status!(ISEI)!(see!for!more!information!Ganzeboom!&!Treiman!2010).! By!transforming!the!ISCOL08!variable!into!a!number!between!0!and!100!the!prestige! level! of! the! occupation! one! works! in! can! made! into! an! easyLtoLworkLwith! interval! variable.!However,!the!4Ldigit!was!used!as!it!is!more!precise!than!the!2Ldigit!variable! and!units!belonging!to!the!same!minor!and!subLmajor!groups!can!also!have!different! values!in!terms!of!prestige.! ! The!gender!variable!originally!had!a!1!and!2!structure.!This!was!changed!in!a!0!for! male!and!a!1!for!female.!Similarly,!the!having!children!had!a!yes!or!no!setup!and!was! also! changed! in! a! 0! for! no! and! a! 1! for! yes.! For! testing! the! first! hypothesis! an! interaction!variable!was!made!with!gender!and!having!children.!!

!

For! the! three! skill! variables! I! used! the! first! out! of! ten! plausible! values! of! literacy,! numeracy!and!problem!solving.!Numeracy!is!here!the!skill!more!closely!associated! with!men,!while!literacy!can!be!deemed!more!feminine.!While!these!two!skills!were! measured! using! regular! test! questions,! the! problem! solving! plausible! value! was! found! by! testing! the! analytical! and! cognitive! capabilities! of! a! respondent.! For! example,!respondents!were!asked!to!go!to!a!webshop!and!purchase!something.!The! reason! why! only! the! first! plausible! was! taken! is! because! the! ten! plausible! values! correlate! heavily! (at! least! 0.85),! and! the! values! are! more! or! less! the! same.! These!

three! skill! variables! were! also! interacted! with! gender! for! answering! their! corresponding! hypotheses.! Initial! results! using! these! plausible! values! were! disappointing.! More! specifically,! the! effect! of! gender! was! not! significant! in! the! models! that! included! the! interaction! effects! for! the! skills! variables.! Furthermore,! because!the!range!of!the!skill!variables!(see!table!1!above)!is!relatively!large!as!well! as!that!none!of!the!respondents!actually!scored!even!a!bit!near!0,!the!effects!were! quite! small! and! the! interpretation! difficult.! For! this! reason! the! command! egen

standardised skill variable = std(plausible skill

variable).!This!creates!a!standardised!variable,!i.e.!the!mean!is!subtracted!from! every! value! and! the! result! is! subsequently! divided! by! the! standard! deviation.! A! standardised! value! can! be! characterised! by! having! a! mean! of! 0! and! a! standard! deviation!of!1.!Consequently,!a!respondent!scoring!a!value!equal!to!the!mean!and! one!standard!deviation!has!precisely!an!!!value!of!1!for!the!standardised!variable,! meaning! that! effect! found! is! the! expected! increase! or! decrease! of! the! dependent! variable.!

!

The!average!proportion!of!females!in!an!occupation!is!measured!by!using!the!egen! command! as! well.! The! command! egen female occ. proportion =

mean(gender), by(ISCO-08 2 digit variable)! can! be! used! to! aggregate! one! aspect! of! one! variable! to! those! of! another,! and! here! it! was! used! specifically!to!create!a!variable!with!the!average!value!of!gender!for!an!occupation,! which!thus!leads!to!the!average!proportion!of!females!in!an!occupation.!Once!again,! this!variable!was!interacted!with!gender!for!answering!its!corresponding!hypothesis.! The! table! below! provides! the! descriptive! statistics! of! the! variables! used! in! the! models.! To! ensure! that! all! models! included! exactly! the! cases! all! observations! that! include!at!least!one!missing!for!any!of!the!below!variables!were!deleted!from!the! dataset.!Lastly,!a!weekly!working!hours!variable!was!also!used!as!a!control!variable! but!was!found!in!most!of!the!models!to!be!insignificant.!It!was!thus!left!out!entirely.! !

Chapter(4.(Results(

!

Section!4.1!Construction!of!the!models!

!

The! first! model! (the! null! model)! basically! only! provides! the! grand! mean! intercept! and!the!variance!of!its!random!intercept!effect.!Model!1!adds!the!gender,!followed! by!model!2!and!the!control!variables.!Model!3!and!all!odd!models!onwards!(except! model!13)!add!the!important!independent!variable!and!model!4!and!all!even!models! onwards!add!the!important!independent!variable!and!its!interaction!variable.!Model! 13!includes!all!variables.! ! Model"0:"" No!independent!variables! Model"1:"" Gender! Model"2;"" Model!1!+!control!variables! Model"3:"" Model!2!+!Children! Model"4:"" Model!3!+!Children*Gender! Model"5:"" Model!2!+!Literacy!score! Model"6:"" Model!5!+!Literacy!score*Gender! Model"7:!! Model!2!+!Numeracy!score! Model"8:"" Model!7!+!Numeracy!score*Gender! Model"9:"" Model!2!+!Problem!solving!score! Model"10:"" Model!9!+!Problem!solving!score*Gender! Model"11:"" Model!2!+!Occupational!female!proportion! Model"12:"" Model!11!+!Occupational!female!proportion*Gender! Model"13:"" All!independent!variables!including!interaction!terms! ! ! Section!4.2!Close!interpretation!of!the!results!

The! regression! results! can! be! found! in! table! 3! on! pages! 24! &! 25.! The! null! model! shows!that!34.2%!of!the!variance!in!hourly!wages!is!simply!explained!by!the!fact!that! individuals!work!in!different!occupations.!This!does!not!change!in!model!1,!although! this! model! does! show! that! women! are! predicted! to! earn! less! than! their! male!

colleagues.!The!negative!effect!of!being!a!woman!becomes!slightly!smaller!in!model! 2,!meaning!that!a!part!of!it!might!be!because!women!have!lower!credentials!and/or! work!experience!then!their!male!cohorts.!This!model!also!adds!the!most!important! control!variables,!and!from!here!out!onwards!the!ICC!stays!at!approximately!5%.!This! also!means!that!these!control!variables!for!a!large!part!point!to!reasons!why!some! work!in!certain!occupations!with!specific!wages.!Clearly,!an!individual!needs!to!have! a! university! degree! to! become! a! doctor,! and! a! high! ranking! professional! needs! to! have!some!degree!of!work!experience!to!attain!that!position.!

!

Adding!the!children!variable!in!model!3!slightly!increases!the!negative!coefficient!for! women! and! the! effect! of! having! children! is! significant! and! positive.! The! reason! behind!the!latter!effect!is!unclear.!After!testing!with!several!other!control!variables! no! explanation! of! multicollinearity! could! be! found.! However,! it! can! be! found! here! that! having! children! seriously! imparts! women’s! hourly! wages! when! compared! to! men’s.!Still,!it!also!seems!that!women!with!children!earn!more!than!women!without.! The!reason!behind!this!is!unclear.!After!testing!with!several!other!control!variables! no!explanation!of!multicollinearity!could!be!found.!Furthermore,!the!model!already! included! work! experience! as! a! control! variable.! In! model! 4,! women! with! children! approximately! earn! only! 91.3%! of! what! men! with! children! earn! every! hour.! Consequently,! hypothesis! 1! cannot! be! rejected,! although! the! data! seems! to! point! out!that!both!women!and!men!in!the!same!occupation!earn!a!premium!when!they! become!parents.!Still,!this!is!less!for!women.!

!

Models!5!to!10!include!the!skill!variables.!In!principle,!all!models!show!that!having! higher!levels!of!numeracy,!literacy!and!problem!solving!skills!is!always!positive!for! one’s! wages.! Interestingly,! the! models! with! standardised! scores! have! significant! coefficients!for!gender!in!all!the!models,!while!that!without!did!not!(see!table!6!in! appendix!1).!However,!only!numeracy!has!a!significant!interaction!effect!in!the!table! presented! in! this! chapter.! This! coefficient! being! negative! means! that! women! are! rewarded! less! for! their! mathematical! skills! than! men,! almost! even! negating! the! positive!effect!of!numeracy!at!all.!More!specifically,!a!women!with!a!numeracy!score!! !

! ! ! ! ! ! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !

of!334.066,!which!coincides!with!an!!!value!of!1!for!the!standardised!plausible!value! as!it!is!the!mean!numeracy!score!plus!1!standard!deviation,!would!earn!only!92%!of! the!hourly!wage!a!comparable!man!in!the!same!occupation!would!get.!This!cannot! be! found! with! the! other! two! skill! scores! because! the! interaction! effect! is! not! significant.! However,! women! still! earn! less! than! their! male! cohorts! in! the! same! occupation!due!to!the!coefficient!of!gender!still!being!positive.!Consequently,!only! hypothesis!3!cannot!be!rejected!while!2!and!4!do!have!to!be.!!

!

Model!11!and!12!provide!evidence!that!the!average!wages!in!an!occupation!do!not! depend! on! the! proportion! of! women! working! in! said! occupation.! However,! what! does! seem! the! case! is! that! women,! when! controlling! for! the! other! variables,! earn! more!than!men!in!the!occupations!with!almost!only!women!(i.e.!when!the!average! proportion!of!women!in!an!occupation!is!near!1).!A!value!of!1!disables!comparing! that!of!men!and!women,!as!men!do!not!work!in!that!occupation!then.!In!this!case,! the!interaction!effect!of!women!working!in!a!highly!feminised!occupation!is!larger! than!the!decrease!they!have!because!they!are!women.!However,!in!the!hypothetical! case!that!only!women!work!in!an!occupation,!they!only!earn!0.5%!more!than!their! (non[existing)!male!cohorts.!This!difference!is!incredibly!tiny,!meaning!that!women! higher!hourly!wage!advantage!over!men!is!practically!non[existant.!In!the!case!of!the! mean! of! average! female! proportion! in! an! occupation,! which! is! 0.497,! women! still! earn!6.7%!less!than!their!male!colleagues.!Interestingly,!this!effect!is!still!present!in! model!13!(although!the!evidence!is!less!strong;!it!can!only!be!said!with!90%!security).! Here!women!earn!approximately!4.1%!less!than!their!male!colleagues!in!the!same! occupation.!Consequently,!both!hypotheses!5!and!6!are!rejected.!Peculiarly,!in!the! case!of!hypothesis!6!the!opposite!is!actually!true,!although!it!must!be!said!that!men! are!unaffected!by!the!feminisation!of!an!occupation.!Moreover,!because!women!still! retain!a!disadvantage!simply!because!of!their!sex,!men!still!tend!to!earn!more!than! women.!All!other!interaction!effects!do!not!show!significant!effects.! ! This!particular!model!is!most!probably!over!fitted,!seeing!how!many!of!the!variables! are!insignificant.!The!VIFs!of!the!model!without!standardised!scores!are!also!abysmal! (see! table! 5! in! appendix! 1).! The! square! root! of! a! variance! inflation! factor! of! an!

independent! variable! denotes! how! much! larger! the! standard! error! is! than! what! it! would! be! in! the! situation! when! that! specific! independent! variable! would! be! unrelated! to! other! independent! variables! (Allison! 1999:! 142).! Some! (e.g.! Allison! 1999)! find! a! VIF! higher! than! 2.5! problematic,! but! others! accept! a! value! of! 5.0.! Because! the! VIFs! of! the! model! without! standardised! skill! variables! were! that! bad! (e.g.!gender!had!a!VIF!of!84.70),!the!skill!variables!had!to!be!standardised!and!the! model!thus!revised.!This!in!turn!ensured!that!effect!of!gender!was!significant!in!all! models.! Even! with! how! over! fitted! this! model! is,! after! controlling! for! all! other! variables!a!woman!is!expected!to!earn!10.2%!less!than!men!in!the!same!occupation.!! Furthermore,!the!ICC!stays!relatively!stable!after!adding!the!control!variables.!Why! this! figure! is! lower! than! it! was! in! the! earlier! models! is! probably! because! the! ISEI! (itself! being! a! derivative! of! the! ISCO[08! variable)! takes! away! a! fair! amount! of! the! occupational! variance.! Lastly,! it! is! interesting! that! model! 4,! which! includes! the! children! variable! and! its! interaction! effect,! has! the! lowest! absolute! log! likelihood! figure! of! all! the! models! except! model! 13.! This! suggests! that! the! effect! of! motherhood! is! more! important! than! the! tested! skills! and! average! occupational! proportion!of!women!for!understanding!gender!hourly!wage!differences.!

Chapter(5.(Discussion( ! Section(5.1(From(results(back(to(theory( ! But!what!do!the!results!mean!for!hypotheses!and!the!theories!behind!them?!Clearly,! in!some!cases!the!hypotheses!simply!had!to!be!rejected,!although!this!is!not!the!case! the! first! one.! This! hypothesis,! in! principle! about! the! existence! of! the! so>called! ‘motherhood!wage!penalty’,!seems!to!be!true!in!the!Netherlands!of!2012.!Still,!there! are! several! aspects! that! need! clarification! in! order! to! understand! what! the! underlying!explanation!is!of!this!gender!difference.!Most!importantly,!are!mothers! basically! discriminated! by! employers,! or! are! they! simply! not! as! good! as! men?! Another! explanation! could! be! that! they! lack! experience! when! compared! to! men! because!mothers!have!left!their!work!momentarily.!This!last!explanation!might!be! the! case,! as! it! can! be! seen! in! table! 3! that! the! effect! of! work! experience! changes! between!model!2!and!3!from!0.040!to!0.034.!The!same!change!can!also!be!seen!in! models!4!and!13.!Although!no!conclusions!can!be!taken!now!from!this,!it!might!be! worthwhile!to!research!this!more!thoroughly.!Interestingly,!the!main!effect!of!having! children! is! also! significant! and! positive! in! all! models! that! includes! it.! The! reason! behind!this!is!unclear!and!this!also!needs!more!attention.!Glauber!also!stated!that! mothers! might! lack! the! human! capital! of! their! male! colleagues! (2012:! 118).! This! thesis! has! not! tested! for! this,! although! it! might! be! interesting! to! do! so.! However,! including!interaction!effects!that!encompass!a!skill,!gender!and!having!children!was! expected! to! be! constraint! by! the! data! as! already! two! of! the! interaction! effects! of! gender!were!not!significant!to!begin!with.!A!further!interesting!point!is!that!the!log! likelihood!was!absolutely!lower!in!the!models!including!the!children!variables.!This! could!mean!that!having!children!or!not!is!an!important!element!for!understanding! the!variance!of!hourly!wages.! ! Other!elements!of!the!results!to!focus!on!are!those!revolving!the!skill!variables.!Only! the! third! hypothesis! about! numeracy! skills! was! actually! proven! by! this! research.! Literacy!skills!were!expected!to!lower!the!wages!of!both!men!and!women!and!this! effect! was! to! be! stronger! for! women.! However,! having! literacy! skills! actually!

improves! the! wages! of! both! men! and! women! (i.e.! the! interaction! effect! is! not! significant).!Similarly,!the!main!effect!for!problem!solving!skills!also!turned!out!to!be! significant! and! positive! for! both! men! and! women,! while! its! interaction! was! once! again! insignificant.! This! could! mean! that! the! human! capital! theory! is! right:! having! more!skills!increases!earnings.!However,!three!aspects!of!the!models!would!oppose! this!conclusion.!Firstly,!according!to!human!capital!theory,!the!gender!wage!gap!is!a! result!of!women!having!fewer!skills!than!men.!This!means!that!the!negative!effect!of! gender! would! disappear! when! controlling! adequately! for! skills! and! this! is! not! the! case! here.! However,! it! might! be! the! case! that! other! variables! would! be! better! substitutes! for! human! capital.! Secondly,! being! a! woman! actually! diminishes! the! return!of!having!numeracy!skills.!In!this!sense,!it!could!be!expected!that!the!theory! of! cultural! valuation! as! explained! by! Kilbourne! et! al.! (1994)! is! actually! right! for! numeracy.!Thirdly,!the!skills!measured!in!the!PIAAC!dataset!can!be!considered!very! broad.! The! fact! that! these! three,! in! general,! work! out! positive! for! both! men! and! women,!has!probably!more!to!do!with!the!fact!that!they!can!be!applied!in!almost! any!occupation.!Moreover,!the!reason!why!all!the!skill!variables!and!their!interaction! effects!are!insignificant!in!model!13!is!because!they!tend!to!be!highly!correlated.!This! is!not!strange!considering!the!fact!that!skills!are!typically!general!in!nature,!as!well!as! that!smart!people!are!not!only!smart!in!one!aspect!of!skills.! !

Also! unexpected! was! the! effect! of! average! feminisation! of! an! occupation.! Firstly,! hypothesis!5!stated!that!working!in!typically!female!occupation!would!lower!average! wages,! but! this! effect! does! not! exist! in! the! Netherlands.! However,! even! after! controlling! for! this! it! seems! that! being! women! still! means! earning! a! lower! hourly! wage! than! men! would! get.! Hypothesis! 6! expected! that! men! would! get! a! wage! premium! in! feminised! occupation! because! they! would! be! scarce! in! them.! In! the! Netherlands,! however,! the! opposite! can! be! found.! In! this! sense,! the! theory! of! tokenism!as!presented!by!Kanter!(1977)!fits!the!Netherlands!better.!Yet!the!negative! effect! of! being! a! woman! still! persists,! and! consequently,! men! earn! more! than! women!in!the!same!occupation!except!in!highly!feminised!occupations.!Even!so,!the! premium!women!earn!in!those!occupations!is!extremely!small!(approximately!0.5%! more).!In!principle,!this!means!that!men!will!mostly!be!better!off.!It!might!be!the!

case!that!the!egalitarian!wage!structure!as!described!by!Mandel!&!Semyonov!(2005)! are! the! reason! for! this,! but! this! is! not! clear.! The! reason! that! the! normal! effect! of! gender!persists!might!have!to!do!that!women!do!not!get!their!yearly!smaller!raises!in! pay!while!men!do.!The!fact!that!the!main!effect!of!gender!is!present!in!all!the!models! might! also! mean! that! there! are! other,! perhaps! also! better! explanations! for! understanding!the!gender!wage!gap!in!occupations.!

! !

Chapter(6.(Conclusion( !

Section(6.1(Conclusion( (

The! research! question! introduced! in! the! beginning! of! this! thesis! related! to! why! Dutch!women!earn!less!on!an!hourly!basis!than!their!male!colleagues.!The!foremost! conclusion!to!be!taken!is!that!Dutch!female!workers!in!nearly!all!situations!earn!less! per!hour!than!their!male!cohorts.!!More!specifically,!it!was!found!that!motherhood! has! a! negative! influence! on! this! hourly! pay! gap.! Although! it! was! not! found! that! literacy! skills! and! problem! solving! skills! negatively! influences! this! gap,! men! do! get! higher!wages!than!women!with!comparable!scores!of!numeracy.!Lastly,!working!in!a! more!feminine!occupation!was!not!found!to!have!an!effect!on!the!average!wages!in! an! occupation! whatsoever,! and! women! tend! to! make! the! pay! gap! between! them! and! men! smaller! in! highly! feminised! occupations.! The! results! not! always! corresponded!with!the!expectations,!meaning!that!it!might!be!fruitful!to!investigate! the!Dutch!more!thoroughly.! ! ! Section(6.2(Recommendations(for(future(studies( ! The!research!in!this!thesis!can!be!seen!as!a!good!introduction!to!understanding!the! Dutch! gender! wage! gap! in! occupations.! However,! there! are! several! issues! that! should!be!addressed!in!due!time.!Firstly,!gender!inequality!comes!in!many!different! forms.!What!is!very!closely!related!to!the!gender!wage!gap!is!the!occupational!sex! segregation.! Fact! is! that! women! and! men! manifest! themselves! in! different! professions.!An!important!aspect!to!look!at!is,!for!example,!the!reason!why!women! tend! to! be! overrepresented! in! certain! occupations,! while! almost! being! invisible! in! another.!Moreover,!why!is!women’s!work!generally!valued!less!than!that!of!men?! Another!interesting!point!is!the!role!of!partFtime!work!for!the!earnings!of!women!as! this!is!a!typical!characteristic!of!the!Dutch!labour!market.!

To!truly!understand!the!problems!introduced!by!motherhood,!more!research!is!also! needed!here.!For!instance,!Glauber!(2012)!divided!having!children!into!having!one,! two!or!three!or!more!children.!She!found!in!her!study!that!the!income!loss!changes! with!successive!childbirths.!Furthermore,!the!social!policies!with!regards!to!childcare! might! change! in! the! coming! years.! Dutch! deputy! prime! minister! and! minister! of! social! affairs! and! employment! Lodewijk! Asscher! has! announced! that! he! is! a! proponent! of! expanding! the! childcare! possibilities! for! fathers! (NU.nl! 2015a),! although!it!is!also!stated!that!only!25%!of!fathers!is!actually!making!use!of!it!at!the! moment!(NU.nl!2015b).!It!is!expected!that!mothers!would!face!less!disadvantages!at! work! when! their! husbands! would! take! similar! times! off,! but! in! order! to! let! this! happen!it!should!be!known!why!fathers!tend!to!decline!their!rights.! ! Moreover,!it!can!be!expected!that!skills!have!some!sort!of!relation!to!earnings,!but! how!exactly!is!not!clear!after!using!the!2012!PIAAC!dataset.!Last!year!I!used!average! work!tasks!in!an!occupation!to!find!an!answer!to!this,!and!this!year!actual!measured,! but!highly!general!skills!were!used.!More!specific!and!actually!measured!skills!might! be! better.! Another! problem! with! the! PIAAC! dataset! is! that! the! actual! number! of! observations!remaining!after!data!cleaning!is!fairly!low.!Of!the!approximately!5,100! respondents!more!than!40%!fell!off!due!to!selection.!The!representativeness!of!the! 5,100!can!surely!be!guaranteed,!but!it!remains!the!question!whether!results!of!those! used!in!this!analysis!can!be!found!again!in!another!dataset.!In!that!sense,!comparing! the!results!of!this!study!with!data!of!a!future!leg!of!the!PIAAC!survey!might!be!very! interesting.! ! ! Section(6.3(Reflection( ! After!focusing!a!large!part!of!studies!on!gender!issues!as!well!as!writing!two!theses! on! the! Dutch! gender! wage! gap,! I! believe! I! made! a! proper! start! on! understanding! gender!inequalities.!While!I!sometimes!faced!difficulties!in!understanding!what!I!was! doing! in! terms! of! methodology! as! well! as! in! substance,! I! feel! that! what! I! did! increased!my!personal!knowledge!about!the!subject.!However,!I!regret!that!I!did!not!

have!the!time!to!look!at!other!gender!inequalities!as!well,!e.g.!the!occupational!sex! segregation.!Some!things!could!have!gone!better,!or!at!least!finished!more!quickly.! Especially!the!theoretical!part!was!difficult!to!write,!probably!more!so!because!of!the! abundance!already!written!about!it.!Even!so,!I!believe!that!this!thesis,!as!well!as!my! studies,!has!helped!me!hone!my!skills!as!a!sociologist!interested!in!gender!issues!and! inequalities.! !

!

References( (

• Agresti,! A.! &! B.! Finlay! (2009)! Statistical( Methods( for( the( Social( Sciences.! Fourth!edition.!Upper!Saddle!River:!Pearson!Education.!

• Allison,!P.D.!(1999)!Multiple(Regression:(A(Primer.!Thousand!Oaks:!Pine!Forge! Press.!

• Budig,! M.J.! &! P.! England! (2001)! ‘The! Wage! Penalty! for! Motherhood’.!

American(Sociological(Review(66:!204Q25.!

• Centraal!Bureau!voor!de!Statistiek!(2007)!Jonge(vrouwen(hoger(opgeleid(dan(

mannen.! Available! through! <http://www.cbs.nl/nr/exeres/20F7E2BCQ6E02Q

4F71QA668Q2AEC94FA2EF3.htm>![1!July!2015].!

• Cohen,!P.N.!&!M.L.!Huffman!(2003)!‘Individuals,!Jobs,!and!Labor!Markets:!The! Devaluation!of!Women’s!Work’.!American(Sociological(Review!68!(3):!443Q63.! • Duyvendak,! J.! &! M.! Stavenuiter! (2004)! ‘Working! Fathers,! Caring! Men:!

Reconciliation!of!Working!Life!and!Family!Life’.!The!Hague!/!Utrecht:!Dutch! Ministry!of!Social!Affairs!and!Employment,!Department!for!the!CoQordination! of!Emancipation!Policy.!

• EspingQAndersen,! G.! (1990)! The( Three( Worlds( of( Welfare( Capitalism.! Cambridge:!Polity!Press.!!!

• EstevezQAbe,! M.,! T.! Iversen! &! D.! Soskice! (2001)! ‘Social! Protection! and! the! Formation! of! Skills:! A! Reinterpretation! of! the! Welfare! State’! In! Varieties( of(

Capitalism.(The(Institutional(Foundations(of(Comparative(Advantage.!Ed.!by!

Hall!P.A!&!D.!Soskice.!Oxford:(Oxford!University!Press:!145Q83.!

• EstévezQAbe,! M.! (2005)! ‘Gender! Bias! in! Skills! and! Social! !Policies:! The! Varieties!of!Capitalism!Perspective!on!Sex!Segregation’.!Social(Politics(12!(2):! 180Q215.!!!

• Ganzeboom,! H.B.G.! &! D.J.! Treiman! (2010)! “International! Stratification! and! Mobility!File:!Conversion!Tools”.!Amsterdam:!Department!of!Social!Research! Methodology.! <http://www.harryganzeboom.nl/ISCO08/index.htm>! [1! July! 2015].!

• Glauber,! R.! (2012)! ‘Women’s! Work! and! Working! Conditions:! Are! Mothers! Compensated!for!Lost!Wages?’.!Work(and(Occupations!39!(2):!115Q38.!