Coworker Interactions in Labor Supply

Rafael Lalive

Description

Coworker Interactions in Labor Supply? Rafael Lalive and Pierpaolo Parrotta† Abstract This paper investigates the role of interactions in labor supply among pairs of coworkers. Coworkers may affect each other since they exchange information on retirement planning through work and leisure interactions. We study non-employment choices among pairs of differently aged coworkers that commute to work and live in the same municipality in Denmark. Identification is based on the fact that workers aged 60 years or older have access to early retirement benefits. We find that as the older coworker' crosses the early retirement age threshold, the younger coworker significantly decreases labor supply. Moreover, when we analyze a reform that increases the financial penalty for early retirement, we find that the effect of a coworker's retirement on own non-employment behavior is substantially reduced. This evidence is consistent with the increase in the financial cost of retiring as the coworker retires. JEL Classification: J26, J14, C40, D10. Keywords: Coworkers' joint retirement, friends at the workplace, work/leisure complementarity. ?We thank Anna Piil Damm, Tor Eriksson, Jakob Roland Munch, Michael Rosholm, Ott Toomet, Olof Åslund (alphabetical order) and from participants at the EALE 2011 and ESPE 2011 conferences for helpful comments and suggestions. Pierpaolo Parrotta acknowledges the financial support from the Graduate School for Integration, Production and Welfare.

• pension system


• retirement


• phenomena5 involving social


• studies concerning social


• observing others' plans


• between coworkers


• coworkers' retirement


• female labor


• rate residence-based national

Sujets

• Aarhus University
• Pension system
• Retirement

Informations

• Publié par: pefav
• Nombre de lectures: 9
• Langue: English

Extrait

CoworkerInteractionsinLaborSupply ∗ RafaelLaliveandPierpaoloParrotta † Abstract Thispaperinvestigatestheroleofinteractionsinlaborsupplyamongpairsofcoworkers.Coworkers mayaffecteachothersincetheyexchangeinformationonretirementplanningthroughworkandleisure interactions.Westudynon-employmentchoicesamongpairsofdifferentlyagedcoworkersthatcommute toworkandliveinthesamemunicipalityinDenmark.Identificationisbasedonthefactthatworkers aged60yearsorolderhaveaccesstoearlyretirementbenefits.Wefindthatastheoldercoworker’crosses theearlyretirementagethreshold,theyoungercoworkersignificantlydecreaseslaborsupply.Moreover, whenweanalyzeareformthatincreasesthefinancialpenaltyforearlyretirement,wefindthattheeffect ofacoworker’sretirementonownnon-employmentbehaviorissubstantiallyreduced.Thisevidenceis consistentwiththeincreaseinthefinancialcostofretiringasthecoworkerretires. JELClassification:J26,J14,C40,D10. Keywords: Coworkers’jointretirement,friendsattheworkplace,work/leisurecomplementarity. ∗ WethankAnnaPiilDamm,TorEriksson,JakobRolandMunch,MichaelRosholm,OttToomet,OlofÅslund(alphabetical order)andfromparticipantsattheEALE2011andESPE2011conferencesforhelpfulcommentsandsuggestions.Pierpaolo ParrottaacknowledgesthefinancialsupportfromtheGraduateSchoolforIntegration,ProductionandWelfare.Theusual disclaimerapplies. † RafaelLaliveisaffiliatedwithUniversityofLausanne,DepartmentofEconomics,CH-1015Lausanne,Switzerland,and CEPR,CESifo,andIZA.E-mail:[e-mail];PierpaoloParrottaisaffiliatedwithAarhusSchoolofBusinessand SocialSciences,AarhusUniversity,DepartmentofEconomics,Hermodsvej22,DK,8230Aabyhoj,Denmark,andUniversityof Lausanne,DepartmentofEconomics,CH-1015Lausanne,Switzerland.E-mail:[e-mail]. 1

1Introduction Theimportanceoftheemployees’retirementchoiceshasbeengrowingduringthelastdecades,notonlyin theeconomicliterature.Thereasonsbehindthisincreasinginterestaremainlyassociatedwiththesweeping changesexperiencedinthelabormarketsoftheOECDcountries: 1 agingpopulation,dropinfertilityrates, increaseinfemalelaborsupplyandretirementrates,andimprovementinthegeneralleveloflivingconditions. Eachofthesefactorshascontributedtomovethesustainabilityofthesocialsecurityandpensionsystemto oneofthetopprioritiesofthepublicpolicies. Thus,thechangeintheratiobetweenworkingandretiredpeoplehaspushedseveralnationalauthorities toreformtheirsocialsecuritylegislationinordertoprovideincentivesaimedatpostponingtheactualworkers’ retirementages.Whereasinseveralcasesthepensionreformsmainlyresultedinaraiseofthestatutory retirementage(especiallyforwomen), 2 theDanishgovernmentsharpenedthecostsofearlyretirementin 1999andloweredtheformalretirementagein2004(from67to65forbothmenandwomen).Thegoalof thisreformwastheshorteningofthegapbetweenstatutoryandactualretirementage. UsingtheDanishlinkedemployer-employeedatabase(LEED)fortheperiod1980-2006,thispaperana- lyzesthedecisionsofjointexitfromemploymentinpairsofcoworkers:itinvestigateswhethercoworkers’ laborsupplydecisionsareaffectedbypeerbehaviorintheworkplace.Moreover,thisstudyevaluateshow thechangeintheincentiveschemein1999affectedthephenomenonunderanalysis. ThedatasetextractedfromtheLEEDiscomposedofallindividualswhohavebeenexclusivepairs ofcoworkersaged50-70,beingnativesandoneolderthantheother,livinginthesamemunicipalityand commutingtowork. 3 Theimplementedestimationstrategyisbasedonaregressiondiscontinuitydesign(RDD)modelandan instrumentalvariable(IV)approachaimedatidentifyingthe(causal)effectofinterestanddealingwiththe typicalendogeneityprobleminvolvedintheso-calledpeereffectanalysis.Specifically,theoldercoworker’s earlyretirementisdirectlyproxiedorinstrumentedinatwo-stagesettingbyusingthetypical(early)retire- mentagerecordedinDenmark.Theinstrumentalvariableapproachallowsustoidentifytheweightedlocal averagetreatmenteffect(LATE)oftheolderretiredcoworkerontheyoungerone’slaborsupply. Findingsfromthisempiricalanalysiswidelysupportthepresenceofjointnon-emploment(earlyretire- ment)intheselectedpairsofcoworkers.Theoldercoworker’sretirementincreasesbyabout9.8percentage pointstheyoungercoworker’sprobabilityofexitemploymentovertheentiretimeperiod,theeffectreaches 17.9percentagepointsbeforetheintroductionofthepensionreformin1999anddecrasesto5.8percentage 1 BlöndalandScarpetta(1998);OECD(1995,2005,2009). 2 Whitehouseetal.(2009). 3 Itmeansthatgroupsofcoworkersfulfillingthementionedconditionsbutcomprisingmorethan2componentshavebeen keepoutoftheanalysis. 2