277

Outsourcing,OccupationalRestructuring,and Employee

Well-beingIs There a Silver

Lining?

Petri BöckermanMika Maliranta

PALKANSAAJIEN TUTKIMUSLAITOS • TYÖPAPEREITA

LABOUR INSTITUTE FOR ECONOMIC RESEARCH • DISCUSSION PAPERS

Helsinki 2012

277Outsourcing, Occupational Restructuring, and Employee Well-beingIs There a Silver Lining?

Petri Böckerman*Mika Maliranta**

This study was funded by the Finnish Work Environment Fund. This paper is also part of the project (No. 134057) financed by the Academy of Finland. Data construction and decomposition computations were

conducted at Statistics Finland, following the terms and conditions of confidentiality. To obtain access to the data, please contact the Research Laboratory of the Business Structures Unit, Statistics Finland, FI-00022,

Finland. We are grateful to Antti Kauhanen for very useful comments. The usual disclaimer applies.

* Labour Institute for Economic Research, petri.bockerman@labour.fi** ETLA – The Research Institute of the Finnish Economy, mika.maliranta@etla.fi

An identical version has been published in the Research Institute of the Finnish Economy discussion paper series (#1271).

TYÖPAPEREITA 277DISCUSSION PAPERS 277

Palkansaajien tutkimuslaitosPitkänsillanranta 3 A00530 HelsinkiPuh. 09−2535 7330Faksi 09−2535 7332www.labour.fi

ISBN 978–952–209–103–1 (pdf)ISSN 1795–1801 (pdf)

Tiivistelmä

Tutkimuksessa tarkastellaan toimintojen ulkoistamisen ja tehtävärakenteiden muutosten vaiku-

tuksia työntekijöiden itsensä kokemiin työoloihin. Työoloja mitattiin useilla eri mittareilla. Tulok-

set perustuvat poikkeuksellisen rikkaaseen yhdistettyyn aineistoon. Tarkastelu osoittaa, että ul-

koistaminen ulkomaille lisää työtehtävien häviämistä yrityksissä. Tämän ulkomaille ulkoistamisen

suoran vaikutuksen lisäksi eriteltiin myös ulkoistamisen epäsuoria vaikutuksia niihin työntekijöi-

hin, jotka onnistuvat säilyttämään työpaikkansa sellaisessa yrityksessä, joka on ulkoistanut toi-

mintojaan. Ulkoistamisen epäsuorat vaikutukset ovat tulosten valossa suoria vaikutuksia huomat-

tavasti heikompia. Tämän lisäksi epäsuorat vaikutukset työntekijöiden kokemaan hyvinvointiin

eivät ole yksioikoisen negatiivisia. Ulkoistaminen toisiin teollisuusmaihin esim. parantaa selvästi

suomalaisten työntekijöiden mahdollisuuksia edetä urallaan.

1Outsourcing, Occupational Restructuring, and Employee Well-being: Is There a Silver Lining?

Contents

Abstract 2

1 Introduction 3

2 Conceptualframework 4

3 Data 8

4 Empiricalspecifications 11

5 Results 12

6 Conclusions 20

Appendix 22

References 26

ETLA Keskusteluaiheita – Discussion Papers No 12712

Abstract

This paper explores the effects of outsourcing on employee well-being through the use of the Finnish linked employer-employee data. The direct negative effect of outsourcing is attributable to greater job de-struction and worker outflow. In terms of perceived well-being, the winners in international outsourcing are those who are capable of performing interactive tasks (i.e., managers, professionals and experts), es-pecially when offshoring involves closer connections to other developed countries.

Key words: globalization, outsourcing, offshoring, working conditions, job satisfaction, subjective well-being

JEL: J28, F23

3Outsourcing, Occupational Restructuring, and Employee Well-being: Is There a Silver Lining?

1 IntroductionOutsourcing, especially international outsourcing (i.e., offshoring), has become the prima-ry method that firms use to change their production structure across developed countries.Blinder(2006)claimsthatoffshoringconstitutesthenextindustrialrevolution.Maloneetal.(2011)arguethattheworkindevelopedcountrieswillbe‘atomized’intoever-smallerpieces.Offshoringhasalreadyledtothesubstantialverticalfragmentationofproduction.Lindenetal.(2007)describethisprocessinthecaseoftheproductionofApple’siPod,andAli-Yrkköetal.(2011)discussitinthecontextofthevaluechainofaNokiasmartphone.Offshoringislike-ly to entail occupational restructuring within firms. Consequently, offshoring increases thepaceofjobdestructionandworkeroutflow,atleastforthosetypesoftasksthatcanbeeasilyandprofitablyoutsourced.Outsourcing, ingeneral,andoffshoring, inparticular,canthere-forebeexpectedtohavenegativeeffectsonperceivedemployeewell-being,especiallyamongemployeeswhoperform“offshorable”tasks.However,evidenceoftheeffectsofoutsourcingonwell-beingamongemployeeswhoarecapableofkeepingtheirjobs(i.e.,stayers)issparse,despitetheimportanceofongoingchanges.11

Thispapercontributestotheliteratureinmultipleways.First,weuseacomprehensivesetoffirm-levelmeasuresofoffshoring,whereasmuchoftheearlierliteraturehasdependedonin-dustry-levelproxiesforoffshoring.Macrodatamaysufferconsiderablyfromaggregationbias,whichhinderstheidentificationoftheeffects(Geishecker2008).Second,usingthefirm-leveldata,wedistinguishbetweendifferentformsofoutsourcingandexaminetheireffectsindif-ferentdimensions.Wealsoanalyzetheeffectsofinsourcing(i.e.,theoppositeofoutsourcing).Third,throughtheuseofthelinkedemployer-employeedata(includinginformationonoccu-pationsandvariousindicators),wegaugedifferentaspectsofoccupationalrestructuringwith-infirms,includingthedestructionoftasks,characteristics(i.e.,theinteractiveornon-routinenature)oftasks,andassociatedworkeroutflowsbetweentaskswithinfirms(intra-firmmobil-ity)andbetweenfirms(inter-firmmobility).Inconductingthisanalysis,weidentifythedis-tinctmechanismstriggeredbyoutsourcingthatarerelevantforemployeewell-being.Fourth,using the linkedemployer-employeedata,wemerge the firm-levelmeasuresofoutsourcingandintra-firmoccupationalrestructuringwithahostofemployee-levelindicatorsofwell-be-ing.Ingeneral,theliteraturehasfocusedononlyafewspecificaspectsofwell-being,suchasperceiveduncertainty.However,differentaspectsofwell-beingarelikelytoproducedifferenteffects.Thesubsequentanalysisofmultipleoutcomesprovidesacomprehensivepictureofthepotentialeffects.Fifth,thedatacovertheservicesector.Earlierresearchhasfocusedonman-ufacturing,buttheshareofmanufacturingjobshasdeclinedconsiderablyindevelopedcoun-tries,andthemanufacturingsectormaynotbearepresentativepartoftheeconomy.There-fore,thisextensionoftheresearchallowsustodeterminewhethertheearlierfindingsarespe-cifictomanufacturing.

We analyze the effects of outsourcing on employee well-being in the Finnish context. ThepressuresofglobalizationareparticularlypronouncedinFinlandbecauseitisasmall,openeconomywithahighlevelofwagesandbenefits.Inrecentyears,considerablechangeshaveoccurred inFinland’s tradepatterns.Forexample, the shareofnon-OECDcountries in thetotalFinnishmanufacturingtradeincreasedbyroughlytenpercentagepointsduring1999–

1 See Crinò (2009) and Eriksson (2010) for surveys of the labor market effects of multinational firms, internationalization, and off-shoring. For example, Østhus and Mastekaasa (2010) study the impact of downsizing on sickness absence.

4

2004.Withinthemanufacturingsector,theelectronicsindustryhasrapidlyincreaseditsout-sourcinginthepasttenyears.Furthermore,theFinnishlabormarkethasbeeninastateofcontinuousturbulencefordecades(IlmakunnasandMaliranta2011).Onaverage,morethantenpercentofall jobsintheFinnishbusinesssectorareeliminatedannually,andthistrendhasbeenstable inrecentyears.Worker inflowandoutflowrateshavebeenmore thantwo-fold.Despiteincreasingglobalizationandturbulentlabormarkets,however,accordingtotheEurobarometer,2lifesatisfactioninFinlandhasbeenatahighlevelwithastableorincreasingtrend(seeLehtoandSutela2009).Thisisnottosay,ofcourse,thatglobalization,withitsvari-ousformsandassociatedlabormarketeffects,hasbeenirrelevantforwell-being,especiallyforthosewhoaremostvulnerabletoitsimpacts.However,offshoringmayplayanimportantroleinreshapingoccupationalstructuresinamannerthatnotonlyincreasesthelaborproductiv-ityandcompetitivenessoftheeconomybutmayalsoprovidemoresatisfyingjobcharacteris-tics(e.g.,moreinteractiveandlessroutinetasks)foremployeeswhoareabletokeeptheirjobsintheprocessofrestructuring.

Thenegativeeffectsofoffshoringonemployeewell-beinghavegainedconsiderableattentionintheliteratureand,especially,inpublicdebate.However,broaderconsiderationssuggestthatoffshoring is likely tohaveapositiveeffect forat least tworeasons.First,offshoring ispartoftherestructuringprocessthatinvolvesanincreaseintheshareofhighvalue-addedoccu-pations in theeconomy.Wedocumentdirectevidence for thisbecauseoffshoring increasestheshareofknowledgeworkersintheaffectedfirms.Restructuringalsofuelstheeconomicgrowththatisfoundtoincreasehappinessindevelopedcountries(Sacksetal.2010;Steven-sonandWolfers2008).Second,ourresultspointoutthattheoffshoring-triggeredrestructur-ingmechanismincreasestheshareofoccupationsintheaffectedfirmsthatarenotonlywellpaidbutarealsofulfillingbecausetheperceivedwell-beingderivedfromthemishighinsev-eraldimensions.

Thispaperproceedsas follows.We firstdescribe theconceptual frameworkand the linkeddata.Anoverviewoftheempiricalspecificationsisprovidednext.Theestimationresultsarethenpresentedandasummaryconcludes.

2 Conceptual frameworkDimensions of Outsourcing and Its Links to Employee Well-being. Figure1illustratesthecon-ceptualframeworkoftheanalysis.Theultimateaimistounderstandhowoutsourcingatthefirmlevelaffectsthewell-beingofstayingemployees.Wepayparticularattentiontotheroleofoccupationalreorganizationwithinfirmsasaconveyingmechanismbetweenoutsourcingandemployeewell-being.Furthermore,weemphasizethefactthateachofthethreepartsoftheanalysis–outsourcing,occupational restructuringandwell-being–hasdiversedimen-sionsthatwarrantclosescrutiny.Figure1revealshowthecombineddata(described inde-tailinthenextsection)provideanexceptionalopportunitytoexaminethethreecloselyinter-linkedpartsandtheirmultipledimensions.Richdatasetsonoutsourcingandemployeewell-beingandcarefulmeasurementofthedifferentaspectsofoccupationalrestructuring,usingacomprehensivesetofindicators,enableustodisentanglethemechanismsofoutsourcingandperceivedwell-being.Thisallowsustoexaminewhetherthenegativeeffectsofoutsourcing

2 See http://ec.europa.eu/public_opinion/index_en.htm.

5Outsourcing, Occupational Restructuring, and Employee Well-being: Is There a Silver Lining?

(foremployeeswhoaremaderedundant)mightbeaccompaniedbypositive,counterbalanc-ingimpacts.

Thenegativeeffectsofoutsourcingonemployeewell-beingare evident.Bydefinition,out-sourcingmeansthatcertaintasksareeliminated(i.e.,movedtootherlocalfirmsorabroad);thus,thedemandforthistypeoflaborinthefirmdecreases.Arguably,asubstantialpropor-tionofthenegativeimpactonwell-beingoriginatesfromtheexpectedlossesoffirm-specif-ichumancapital,quasi-rents(aworkerexpectstoearnlessinthefuture)andcostsincurredbythesearchforanewjob.Itshouldalsobenotedthattheworkerswhomanagetokeeptheirjobs might experience a decrease in perceived well-being. First, the decision to outsourcemightindicatefurtheractionsofasimilarsort;thus,therecentoutsourcingmayincreaseun-certaintyabout thefuture.Second,areductioninpersonnelmayentaildecreasedprospectsforpromotion,aweakenedbargainingpositionandvoiceintheorganization,andanincreasedsenseofdiscrimination becauseonlysomeemployeesaresubjecttooutsourcing.Third,out-sourcingisassociatedwithdownsizinginsometasks,withtheresultthatwork intensitymightbehigher.Allofthesecircumstancessuggestthattheexpectedeffectonsatisfactionlevel isnegative.

36

FIGURE 1. DIMENSIONS AND LINKS BETWEEN OUTSOURCING AND

EMPLOYEE WELL-BEING.

DIMENSIONS1.Job satisfaction2.Uncertainty3.Harm4.Hazard5.No promotion6.No voice7.Discrimination8.Work intensity

TYPE1.Domestic outsourcing2.Offshoring3.Insourcing

DESTINATION1.Developed countries2.Developing countries

TYPE1.Task creation and destruction2.Task hiring and separation3.Internal hiring and separation4.Excess task reallocation5.Worker churning6.Change in the share of interactive, non-routine and knowledge tasks

OUTSOURCING WELL-BEINGOCCUPATIONALRESTRUCTURING

Firm-level Individual-level

International Sourcing Survey

(ISS)

Quality of Work Life Survey (QWLS)

The Finnish Longitudinal Employer-

Employee Data (FLEED)

Anal

ysis

Data

DIMENSIONS1.Job satisfaction2.Uncertainty3.Harm4.Hazard5.No promotion6.No voice7.Discrimination8.Work intensity

TYPE1.Domestic outsourcing2.Offshoring3.Insourcing

DESTINATION1.Developed countries2.Developing countries

TYPE1.Task creation and destruction2.Task hiring and separation3.Internal hiring and separation4.Excess task reallocation5.Worker churning6.Change in the share of interactive, non-routine and knowledge tasks

OUTSOURCING WELL-BEINGOCCUPATIONALRESTRUCTURING

Firm-level Individual-level

International Sourcing Survey

(ISS)

Quality of Work Life Survey (QWLS)

The Finnish Longitudinal Employer-

Employee Data (FLEED)

Anal

ysis

Data

Figure 1 Dimensions and links between outsourcing and employee well-being

6

Thelinkbetweenoutsourcingandwell-beingismoreambiguouswhenthevariabilityinout-sourcing,theheterogeneityofemployees,andthedifferentaspectsofwell-beingarefullycon-sidered.Inotherwords,outsourcingislikelytohavediverseeffectsondifferentdimensionsofwell-being.Theeffectsmayalsovarysignificantlybetweendifferenttypesofsurvivingem-ployees because the adjustment does not treat all employees equally. Some employees maygainintheprocess,whileothersmaylose.Thispotentiallyobscuresthegeneralrelationshipbetweenoutsourcingandperceivedwell-being.Forinstance,knowledgeworkersmaybenefitdisproportionately fromoffshoringbecause itcreatesopportunities for themtoutilize theirskills.

Theeffectsmayalsodifferbythegeographicaldestinationofoutsourcing.Survivingemploy-eesmayperceiveoffshoringtodevelopingcountriesasasignofweaknessinthefirm’sposi-tioninthemarket.Moreover,thequalityofjobsisdrasticallylowerindevelopingcountriesthan in Finland, which constitutes a potential threat to domestic labor standards. Indeed,thereissomeprevioussupportfortheheterogeneityoftheeffectsofoffshoring.Geisheckeretal.(2011)reportedthatoutsourcingtohigh-wagecountriessignificantlyimprovestheper-ceivedlevelofjobsecurityamongGermanemployees,butoffshoringtodevelopingcountrieshastheoppositeeffect.Furthermore,theunderlyingmotivationforoffshoringmayhaveim-plicationsforitssubsequentwell-beingeffects.Inparticular,itisreasonabletoarguethatoff-shoringshouldhavemorenegativeeffectsonemployeewell-beingiftheactivityismotivatedbythereductionoflaborcostsratherthanbyopeningnewmarketsforproductsandservic-esthatwouldbenefitboththefirmanditsworkforceinthelongrun.3Thebottomlineofthediscussionaboutthepotentialheterogeneityoftheeffectsisthatthereisanapparentneedtoestimatespecificationsthatallowforflexibleeffects.

Measurement of Occupational Restructuring. Afirmisacollectionofjobswithdifferenttasks.Occupationalrestructuringistheresultoftaskcreationanddestructioninthefirm.Wegaugevariousaspectsofintra-firmoccupationalrestructuringbyapplyingthestandardmeasuresofjobandworkerflowsattheleveloffirmsinsteadofatthelevelofasectororanindustry,asistypicallydoneintheliterature(Burgessetal.2000;DavisandHaltiwanger1999).Tomeas-uretaskcreationanddestruction,weidentifythenumberofworkersindifferenttasksinthefirmusingtheISCO-88classificationofoccupationsatthe1-digitlevel.4Thegroupsareasfollows:

1. Managers2. Professionals3. Techniciansandassociateprofessionals4. Clerks5. Serviceandcareworkersandshopandmarketsalesworkers6. Craftandrelatedtradeworkers7. Plantandmachineoperatorsandassemblers8. Elementaryoccupations.

3 Ali-Yrkkö (2007) has reported that cost savings have been an important motivation behind outsourcing for the Finnish companies.4 Skilled agricultural and fishery workers are excluded from the analysis because we focus solely on the non-farm business sector. Our general approach resembles the approaches of Bauer and Bender (2004) and Askenazy and Moreno Galbis (2007), who also study intra-firm organizational changes.

7Outsourcing, Occupational Restructuring, and Employee Well-being: Is There a Silver Lining?

Taskcreation(TC)infirmi isthesumofpositiveemploymentchangesinthetasks(j=1,…,8)

betweenyeartandt-1, , ,where denotesthedifferenceoperatorandthe

superscript “+” indicates that Lijt > Lij, t–1. Task destruction (TD) is defined analogously:

,wherethesuperscript“-”indicatesthatLijt < Lij, t–1.Thenetemployment

changeinfirmiis .Therefore,afirmmayexperiencesimul-

taneoustaskcreationanddestruction.Followingtheliteratureonjobflows,asuitableindica-torofsuchactionsisexcesstaskreallocation(ETR):ETRit = TCit + TDit – NETit.

Themeasuresofworkerflowsprovideausefulextensiontotheanalysisofoccupationalre-structuring.ItholdsthatNETit = TCit – TDit = Hit – Sit,whereH(hired)denotesthenumberof employees who were hired for the current task in year t, and S (separated) indicates thenumberofemployeeswholefttheirtaskinyeart.Thehiredemployeesconsistoftwogroups:internally hired (IH) employees, who worked for the same firm (but in a different task) inyeart-1,andexternallyhired(EH)employees,whodidnotworkforthesamefirminyeart-1.Analogously,theseparationscanbedividedintointernallyseparated(IS)andexternallysepa-rated(ES).Thus,itholdsthatNETit = TCit – TDit = Hit – Sit = IHit + EHit – ISit – ESit.Bydefini-tion,IHit = ISit.Tomeasuretheamountof“excessive”workerturnoverinthefirm,wecanusethechurningflowmeasure:CFit = Hit + Sit – (TCit + TDit).

Following the convention in the literature on job and worker flows, all flow measures areconverted into ratesbydividing themby theaverageemploymentof the firm inyear t and

t-1(AL); .5 In the empirical analysis,

wedonotuseannualchanges(i.e.,changesbetweent-1andt);instead,weuseasix-yearwin-dow(i.e.,changesbetween2000and2006).Thischoiceisdictatedbythestructureandcon-tentofourdata.Furthermore,longerdifferencesareusefulforcapturingtime-consumingandgradualmechanisms,suchasthoseexaminedinthisstudy,especiallywhenthedatacontainsomeshort-run“noise”(GrilichesandHausman1986).

Inadditiontothemeasuresoftaskflowrates,wealsoapplyindicatorsthatgaugetheshareofinteractiveandnon-routinetasksinthefirms.Bymeasuringthechangesintheseindicatorsbetween2000and2006,wecanexploreinterestingcharacteristicsofoccupationalrestructur-ingatamoredetailedlevel.Thisopportunityexistsbecausetheindicatorsofthesharesofin-teractiveandnon-routinetasksaredefinedbyusingtheISCO-88classificationoftheoccupa-tionsatthe2-digitlevel,followingBeckeretal.(2009).6Non-routinetasksinvolvenon-repet-itiveworkmethodsandcreativeproblemsolving;theycannotbeprogrammedassimplerules.Interactivetasksrequirepersonalinteractionwithco-workersorthirdparties.Thiscategori-zationofdifferentoccupationsintermsoftheircontentisrelatedtothemeasuresofoutsourc-ingbecauseroutineandnon-interactivetasksaremosteasilyoffshored(Baldwin2006;Beck-eretal.2009).Amorestraightforwardmeasureofoccupationalrestructuringconsistsofthechangeintheshareofknowledgeworkersbetweentwopointsintime.Intheempiricalanal-ysis, “knowledge workers” comprise a broad category, including the first four occupational

5 A useful property of using the average employment as a denominator is that the growth rates are symmetric around the zero (Davis and Haltiwanger 1999).6 Nilsson Hakkala et al. (2009) use the same classification of occupations and the same information on the skill content of tasks.

8

1

jit ijtj

TC L= +=

= ∆∑ ∆

8

1

jit ijtj

TD L= −=

= ∆∑8 8

, 11 1

j jit ijt ij tj j

NET L L= =

−= == −∑ ∑

( ) ( )8 8, 1 , 11 1

2 2j jit ijt ij t it i tj j

AL L L L L= =

− −= == + = +∑ ∑

8

groups(i.e.,managers,professionals,techniciansandassociateprofessionals,andclerks)be-cause, inmostworkplaces, techniciansandassociateprofessionalsaswell as clerkswork incloseco-operationwithprofessionals.7

Therefore, in theempirical specifications,weuse the followingelevenmeasuresofoccupa-tionalrestructuring:1)taskdestructionrate,2)taskseparationrate,3)externaltasksepara-tionrate,4)internaltaskseparationrate,whichisequaltotheinternaltaskhiringrate,5)ex-cesstaskreallocationrate,6)churningflowrate,7)thechangeintheshareofinteractivetasks,8)thechangeintheshareofnon-routinetasks,9)taskcreationrate,10)taskhiringrate,and11)thechangeintheshareofknowledgeworkers.

3 DataTheanalysis isbasedonrich, linkeddatathatcombinethreedifferentdatasources(seethebottompanelsofFigure1).Eachsourcehassubstantialmeritsforthestudyoftheeffectsofoutsourcing.

International Sourcing Survey. To measure firms’ outsourcing activities, we use a firm-levelsurvey,theInternationalSourcingSurvey(ISS)ofStatisticsFinland(SF),conductedin2009(seeStatisticsDenmarketal.2008).Thequestionsonthissurveyrefertodomesticoutsourc-ingandoffshoringduringtheperiod2001–2006andcoverthenon-financialbusinesssector(NACE,sectionsCtoIandK).ThefocusoftheISSwasonlargeenterprisesbecausemulti-nationalenterprisesareconsideredkeyplayers,particularlyinoffshoring.Arandomsampleofsmallerfirms(50–99employees)wasalsoanalyzed,butthecoverageofthesurveyonlarg-erfirms(atleast100employees)ismuchmorecomplete.Theresponserateofthesurveywas80%.Thefinaldatacover1,400firms.Approximately300ofthesefirmshaveaworkforceof50–99employees,andotherfirmsinthesurveyhaveat least100employees.Becauseoftheframeworkofthequestionnaire,thedatacoverasubstantialproportionofthetotalemploy-mentintheFinnishbusinesssector.Forexample,inthemanufacturingsector,theISScover-ageis60%.Intheservicesector,thecoverageis46%ofthefirmsthatemployatleast5per-sons(Maliranta2011).

OffshoringisdefinedintheISSasthetotalorpartialmovementofbusinessfunctions(coreorsupportbusinessfunctions)currentlyperformedin-houseordomesticallyoutsourcedbytheresidententerprisetoeithernon-affiliated(externalsuppliers)oraffiliatedenterpriseslocatedabroad(StatisticsDenmarketal.2008,p.13).Alloutsourcingindicatorsusedinthefollowinganalysismeasuretheoutsourcingofafirm’scorebusinessfunctionsbecausetheoutsourcingofthesefunctionsislikelytohaveaneffectonsurvivingemployees’well-being.8TheISSalsoincludesinformationaboutdomesticoutsourcing,thegeographicaldestinationsofoffshoring,thefirms’motivationsforoffshoringandinformationoninsourcing(i.e.,theoppositeofoff-shoring)ofcorebusinessfunctions.

7 Hopp et al. (2009) consider specific aspects of white-collar tasks at the individual, team and organization levels.8 The definition of core business function is the production of final goods or services intended for the market or for third parties that are conducted by the enterprise and yield income. In most cases, the core business function equals the primary activity of the enterprise. It may also include other (secondary) activities if the enterprise considers these to be part of its core functions (Statistics Denmark et al. 2008, p. 13).

9Outsourcing, Occupational Restructuring, and Employee Well-being: Is There a Silver Lining?

Finnish Linked Employer-Employee Data. ThesecondconfigurationofdatathatweuseintheanalysisistheFinnishLongitudinalEmployer-EmployeeData(FLEED).Thesedataarecon-structedfromanumberofdifferentregistersonindividualsandfirmsthataremaintainedbySF.TheycontaininformationfromEmploymentStatistics,whichrecordseachemployee’sem-ployerduringthelastweekofeachyear.FLEEDareprimarilyusedtomeasureoccupationalrestructuringinfirmsusingthemeasuresoftaskflowsproposedbyMaliranta(2009,2011).The measures of occupational restructuring are based on the ISCO-88 classification at the1-digitlevel(Maliranta2009,2011),asdescribedearlier.9Oneimportantadvantageofthesemeasuresisthat,bydesign,theyareabletotakeintoaccounttheintensityofrestructuring.Incontrast,themeasuresofoutsourcingfromtheISSarebinaryindicators.

Quality of Work Life Survey. Thethirdsetofdatathatweuseisthelatestedition(2008)oftheQualityofWorkLifeSurvey(QWLS)ofSFtomeasureemployeewell-being(LehtoandSute-la2009).TheQWLSprovidesarepresentativesampleofFinnishwageandsalaryearners(theself-employedareexcluded).Theinitialsampleforthissurveyisderivedfromamonthlylaborforcesurvey(LFS)bySF,forwhicharandomsampleoftheworking-agepopulationisselectedfortelephoneinterviews.TherepresentativesampleofemployeesintheQWLSprovidesasig-nificantadvantageoverpreviousstudies,whichhavefocusedonafewmanufacturingindus-triesorsinglefirms.Theestimatesforcertainsectorsandfirmscouldbesubjecttoselectionbiasiftheunobservedfactorsthatdeterminewhetheremployeeschoosetoworkinthesectororfirmalsoinfluencetheirperceivedwell-being.

The2008QWLSwasbasedontheLFSrespondentsinMarchandAprilwhowere15–64yearsoldandhadanormalweeklyworkingtimeofatleasttenhours.6,499individualswereselect-edfortheQWLSsampleandinvitedtoparticipateinapersonalface-to-faceinterview.Ofthissample,4,392personsparticipated(approximately68%)(LehtoandSutela2009),whichwasaveryhighresponserate(68%)foracomplexandburdensomeface-to-facesurvey.Theaver-agelengthoftheinterviewswas66minutes.Face-to-faceinterviewsensurereliableanswerstoalmostallquestions.Duetomissing informationonsomevariables forsomeemployees,thefinalsamplesizeoftheQWLSincludedapproximately4,300observations(~30%ofthesecover thepublic sector,which isnot included inouranalysis).TheQWLS is supplementedwithinformationfromtheLFSandseveralregistersmaintainedbySF.Forexample,informa-tionabouttheeducationallevelofemployeesoriginatesfromtheRegisterofCompletedEdu-cationandDegrees.

Weusedseveralvariablestocaptureemployeewell-being,basedontheQWLS.Someofthesemeasuresweregeneralmeasuresofwell-beingatwork,suchas jobsatisfaction,whileothervariablescapturedmorespecificaspectsofemployeewell-being,suchasperceivedworkin-tensity.Jobsatisfactionwasmeasuredonafour-pointLikertscale.Negativejobaspectsweremeasured according to Böckerman and Ilmakunnas (2008). For perceived uncertainty, therespondents stated whether certain aspects were insecurity factors, including the threat oftemporarydismissalandthethreatofunemployment.Perceivedharmswereratedonafive-pointscale,withthehighestcategorycorrespondingtoanemployee’sperceptionthatacertainworkingconditionwas‘verymuch’anadversefactorintheworkplace.Harmsincludedheat,cold,anddust,amongothers.Forperceivedhazards,themostseriousofthreepossibilitieswas

9 Maliranta (2011) provides detailed descriptive evidence on the roles of occupational restructuring in the context of the Finnish business sector.

10

therespondent’sdesignationofaparticularworkplacefeatureas ‘adistincthazard’.Hazardsincludedaccidentrisk,straininjuryrisk,andworkexhaustionrisk,amongothers.Respons-estothequestionsaboutadverseworkingconditionswereaggregatedbycreatingadummyvariableequaltooneiftherewasatleastoneclearlyadversefactor(Harm),adummyvaria-bleequaltooneiftherewasatleastonedistincthazard(Hazard),andadummyvariableforatleastoneinsecurityfactor(Uncertainty).1010Theseformulationswerenotparticularlysensi-tivetopotentialmeasurementerrorintheself-reportedmeasuresofworkingconditions.Fur-thermore,weusedindicatorsforpoorpromotionprospects,lackingavoiceintheworkplace,andexperiencingatleastonetypeofdiscrimination.Finally,wecapturedperceivedworkin-tensitybyusingtherespondent’sagreementwiththestatement,‘Workpressureincreasessick-nessabsence’.

Matching. Matchingthesethreedatasourcesispossiblebecauseallofthedatasetsthatweusecontainthesameuniqueidentifiersforfirmsandpersons,maintainedbySF.Thisinformationalsoensuresnear-perfecttraceabilityofemployersandemployeesovertime.TheQWLSandFLEEDarematchedbyusingtheuniqueIDcodesforpersons.UsingFLEED,wecanfollowtheemployeeswhoparticipatedinthe2008QWLSovertheperiod1990–2007.Ineachyear,wecanlinkinformationonthefirmandtheestablishmenttoeachperson.ThecombinationoftheQWLSandFLEEDcanthenbematchedtotheISSbyusingtheuniquefirmcodes.ThevariablesthatareusedintheempiricalspecificationsaredescribedindetailintheAppendix(TableA1).

TheQWLSisacross-sectionaldatasetthatincludesonlylimitedself-reportedinformationonpastlabormarketexperience.However,becauseFLEEDcanbeusedtoincorporateinforma-tiononemployees’workhistoryovertheperiod1990–2007,weareablemeasurevariouslabormarketoutcomesinthepast.Thisisparticularlyimportantinourcontextbecauseweareuna-bletoestimatespecificationswithindividualfixedeffects.Byusingthevariablesthatdescribepastlabormarketoutcomes,weareabletotakeintoaccountotherwiseunobservabledetermi-nantsofsubjectivewell-being(seeLechnerandWunsch2011,foranapplicationofthisideainanothercontext).Thespecificationsthatweestimateforemployeewell-beingassumethatun-observedheterogeneityisnotcorrelatedwiththeexplanatoryvariableofinteresttoestablishacausaleffect.Intheempiricalspecifications,weusepastaverageearningsandthenumberofemploymentandunemploymentmonthstodescribeemployees’relevantworkhistory.

BecausetheQWLSdataarefrom2008,inthefinalestimationsample,weuseonlythoseem-ployeeswhowereemployedinthesamefirmduringtheperiod2006–2008.Thematcheddatacontain770observations.ThisnumberreflectsthefactthattheISSdataaremuchmorelike-lytopertaintolargefirms.Thefinalestimationsamplecontainsobservationson367firms;therefore,wehave,onaverage,twoobservationsforeachfirm.Wealsoestimateseparatespec-ifications forknowledgeworkers,witha sample sizeof421.The specifications thatuse themeasuresofoccupational restructuringarebasedona largerdata setof1,174observationsbecausewedonothavetorelyonafirm-levelsurvey(ISS).Instead,wecanusecomprehen-siveregisterdatafromFLEEDtoconstructthemeasuresofturbulence.Thenumberofdiffer-entfirmsinthissampleis796.

10 The most common elements of uncertainty are ‘unforeseen changes’, ‘work load increases beyond tolerance’, and ‘transfer to other duties’. These components of uncertainty typically affect the same employees. The perception of the threat of becoming incapable of work is also quite common (25% of all employees). This threat is much more frequent among older employees, as expected. Note that we control for the age effects in all specifications for perceived well-being.

11Outsourcing, Occupational Restructuring, and Employee Well-being: Is There a Silver Lining?

4 Empirical specificationsThe initialstep intheanalysis is toestablishthe impactofoutsourcingonoccupationalre-structuringbyusingfirm-levelregressions(thefirstandsecondpanelsofFigure1).Thesees-timatesrevealthedirecteffectsofoutsourcingonemployees.Thespecificationstakethefol-lowingform:

(1)

whereRESTRUCTURINGjkrepresentsthemeasurekofoccupationalrestructuringforfirmj. Weuseasthedependentvariableselevendifferentmeasuresofoccupationalrestructuring,asdescribedearlier.Xj represents thevectorofcontrolvariables,which includethesizeof thefirm(thelogarithmofemployment)andtheindustryeffects(withasetofindicatorsfor22in-dustries).Thevariableofinterestisthemeasureofoutsourcing.Inallspecifications,thebase-linecategoryisthatthefirmhasneitheroutsourceddomesticallynoroffshoreditscorebusi-nessfunctionsduringtheperiod2001–2006.Thesespecificationsareestimatedbyusingem-ployment-weighedOLS.Withthisapproach,thefirm-levelregressionsofequation(1)canbeinterpretedasdatawithobservedmeansonindividualemployees.Descriptivestatistics(TableA2,Columns1–3)revealthatthedatacontaininformationonfirmsthatcollectivelyemploy~350,000employees,whichisapproximatelyone-fourthofallFinnishprivatesectoremploy-ees.Approximately100,000employeeshaveworkedinfirmsthathavebeensubjecttosometypeofoutsourcing.

Toexaminetheconnectionbetweenoutsourcingandperceivedwell-beingamongsurvivingemployees(thefirstandthirdpanelsofFigure1),weestimatespecificationswiththefollow-ingstructure:

(2)

whereYijkisthemeasurekofemployeewell-beingforindividuali employedinfirmj. Weuseasthedependentvariableseightdifferentmeasuresofemployeewell-being.Xij representsthecontrolvariables,whichincorporatethestandardindividual-levelcovariates,suchasemploy-ees’ageandeducationlevel,basedontheliteratureonsubjectivewell-being(Clark1996).Thestandarderrors inall specificationsofequation(2)areclusteredat the firmlevel.Columns4–5ofTableA2providedescriptivestatisticsfortheemployee-leveldata.

Notethatoutsourcingactivitiescanbetreatedasexogenoustoindividualemployees.Theevi-denceindicatesthatoffshoringfirmsaremoreproductivethannon-offshoringfirms(Wagner2011).Forvariousreasons,weexpectapositiverelationshipbetweenproductivityandwag-esatthefirmlevel,andtheempiricalevidencesupportsthisexpectation(Abowdetal.1999;Baggeretal.2010).Equation(2)canbeinterpretedasatestoftheexistenceofcompensatingwagedifferentialsbecauseoutsourcingcanbeseenasapotentialdisamenityfromtheemploy-ees’pointofview.Itcanbeshownthatthewageanditsdeterminantsshouldnotbeincludedamongtheright-handsidevariablesoftheequationiftheobjectiveistotestfortheexistenceofcompensatingwagedifferentialsbyusinginformationonsubjectivewell-being(seeBöck-ermanetal.2011).Inprinciple,thepresenceofsomesortofbiasinourestimatescannotbefullyruledoutbecause,forinstance,someemployeesmaybeabletoanticipateupcomingout-

, 1,...,11,jk j j jRESTRUCTURING OUTSOURCING kβ δ ε= + + =X

, 1,...,8,ijk ij j ijY OUTSOURCING kβ η ε= + + =X

12

sourcing.However,givenourcontext,thisbiasislikelytobenegligible,andthedirectionisuncertain.

Finally,weexploretheeffectofoccupationalrestructuringonsurvivingemployees’well-be-ing(thesecondandthirdpanelsofFigure1).Forthispurpose,weusespecificationswiththefollowingstructure:

(3)

whereYijk isthemeasurekofemployeewell-beingforindividuali employedinfirmj. Theex-planatoryvariablesofinterestareeachseparatemeasure(l=1,…,11)ofoccupationalrestruc-turing.ThevectorofcontrolvariablesXij isexactlythesameasinequation(2).ThelasttwocolumnsofTableA2documentdescriptivestatisticsforthedatathatareusedwiththesespec-ifications.

5 ResultsOutsourcing and Occupational Restructuring. Wefirstexaminewhethertheoutsourcingactiv-itiesofFinnishfirmsareassociatedwithoccupationalrestructuringand,ifso,inwhatways.Becauseweareparticularlyinterestedintheroleoffirms’outsourcingfromtheperspectiveof employees, the baseline estimates refer to employment-weighted regressions (Table 1).11Anadditionaladvantageoftheemployment-weightedregressionsisthattheyputgreaterem-phasisonlargerfirms,forwhichthemeasuresofoccupationalrestructuringaremorerelia-ble(IlmakunnasandMaliranta2005).TheresultsinTable1refertocontinuingfirmsbecausemostoftherestructuringoccursamongthem.Withthisrestriction,weavoidtheasymmetriescausedbyentriesandexits.Becauseweincludethefullsetofindustryindicatorsamongthecontrolvariables,theresultspointtowithin-industryeffects.

Theestimatesofequation(1)reportedinTable1revealacoherentpattern.Statistically,off-shoringhasaverysignificantpositiverelationshipwiththeratesoftaskdestructionandtaskseparation (PanelA,Columns1–2).The latter connectionprevailsbecauseof external taskseparation(PanelA,Column3).Incontrast,offshoringisnotconnectedtointernaltasksepa-ration(and,thus,neitherisinternaltaskhiring;seePanelA,Column4).Furthermore,theef-fectofoffshoringonthe“creative”sideofoccupationalrestructuringappearstobenegative,asindicatedbythesignificantnegativecoefficientsforthetaskcreationandtaskhiringrates(PanelA,Columns9–10).Thepointestimatesofoffshoringonthesharesofinteractiveandnon-routinetasks(atthe2-digit leveloftheISCO-88classification)arepositive,buttheef-fectsarenotstatisticallysignificant(PanelA,Columns7–8).Ourbroadermeasureofthecom-positionof theworkforcegivessupport to theargument thatoffshoringcontributessignifi-cantlytotheincreaseinknowledgeworkinfirms(PanelA,Column11).

PanelBofTable1providesamoredetailedbreakdownofoutsourcingaccordingtogeograph-icaldestination.WefindthatoffshoringfortherestofEuropeanddevelopingcountrieshasthemostpronouncedpositiveeffectsontaskdestructionandtaskseparation(PanelB,Col-umns1–2).Thesetypesofoffshoringhaveapositiveimpactontheshareofknowledgework

11 We use the average employment in 2000 and 2006.

, 1,...8 1,...11,ijk ij jl ijY RESTRUCTURING k lβ λ ε= + + = =X

13Outsourcing, Occupational Restructuring, and Employee Well-being: Is There a Silver Lining?

37

TAB

LE 1

THE

EFFE

CT

OF

OU

TSO

UR

CIN

G O

N O

CC

UPA

TIO

NA

L R

ESTR

UC

TUR

ING

AT

THE

FIR

M L

EVEL

Pane

l A

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(1

1)

Ta

sk

dest

ruct

ion

Task

se

para

tion

Exte

rnal

ta

sk

sepa

ratio

n

Inte

rnal

task

se

para

tion

(i.e.

, in

tern

al ta

sk

hirin

g)

Exce

ss ta

sk

real

loca

tion

Wor

ker

chur

ning

Th

e ch

ange

in

the

shar

e of

in

tera

ctiv

e ta

sks

The

chan

ge

in th

e sh

are

of n

on-

rout

ine

task

s

Task

cr

eatio

n Ta

sk h

iring

Th

e ch

ange

in

the

shar

e of

kn

owle

dge

wor

kers

D

omes

tic o

utso

urci

ng

-0.0

499

-0.0

223

-0.0

567

0.03

44**

0.

0447

0.

0551

-0

.017

5**

-0.0

155

-0.0

129

0.01

47

0.00

171

(0

.043

6)

(0.0

391)

(0

.037

9)

(0.0

156)

(0

.030

7)

(0.0

433)

(0

.007

45)

(0.0

105)

(0

.057

4)

(0.0

501)

(0

.016

9)

Off

shor

ing

0.19

2***

0.

216*

**

0.19

0***

0.

0267

0.

0074

1 0.

0494

0.

0132

0.

0217

-0

.212

**

-0.1

87**

0.

0443

**

(0

.072

9)

(0.0

646)

(0

.065

3)

(0.0

210)

(0

.040

7)

(0.0

493)

(0

.012

3)

(0.0

187)

(0

.088

1)

(0.0

768)

(0

.021

0)

N

1096

10

96

1096

10

96

1096

10

96

1096

10

96

1096

10

96

1096

Pa

nel B

(1

) (2

) (3

) (4

) (5

) (6

) (7

) (8

) (9

) (1

0)

(11)

Task

de

stru

ctio

n Ta

sk

sepa

ratio

n Ex

tern

al

task

se

para

tion

Inte

rnal

task

se

para

tion

(i.e.

, in

tern

al ta

sk

hirin

g)

Exce

ss ta

sk

real

loca

tion

Wor

ker

chur

ning

Th

e ch

ange

in

the

shar

e of

in

tera

ctiv

e ta

sks

The

chan

ge

in th

e sh

are

of n

on-

rout

ine

task

s

Task

cr

eatio

n Ta

sk h

iring

Th

e ch

ange

in

the

shar

e of

kn

owle

dge

wor

kers

D

omes

tic o

utso

urci

ng

-0.0

361

-0.0

0777

-0

.037

5 0.

0297

* 0.

0276

0.

0567

-0

.017

1**

-0.0

147

-0.0

312

-0.0

0283

0.

0066

6

(0.0

395)

(0

.036

0)

(0.0

343)

(0

.016

1)

(0.0

343)

(0

.043

4)

(0.0

0700

) (0

.009

96)

(0.0

581)

(0

.049

6)

(0.0

161)

O

ffsh

orin

g to

the

15 E

U c

ount

ries

-0.0

938

-0.0

588

-0.0

719

0.01

31

0.08

68

0.07

01

0.00

595

0.00

453

0.01

97

0.05

47

-0.0

0408

(0.0

652)

(0

.060

3)

(0.0

634)

(0

.026

5)

(0.0

610)

(0

.055

8)

(0.0

0988

) (0

.016

5)

(0.0

817)

(0

.069

5)

(0.0

254)

O

ffsh

orin

g to

the

rest

of E

urop

e 0.

241*

**

0.20

0**

0.22

4***

-0

.023

5 -0

.070

1***

-0

.081

1 0.

0142

0.

0219

-0

.111

-0

.151

* 0.

0327

(0.0

879)

(0

.077

6)

(0.0

812)

(0

.022

6)

(0.0

265)

(0

.062

4)

(0.0

167)

(0

.025

9)

(0.0

938)

(0

.080

2)

(0.0

298)

O

ffsh

orin

g to

dev

elop

ing

coun

tries

0.

171*

0.

207*

* 0.

139

0.06

82**

* 0.

0919

**

0.07

18

0.00

334

0.00

587

-0.2

06

-0.1

70

0.05

74**

(0.0

976)

(0

.086

1)

(0.0

924)

(0

.024

6)

(0.0

439)

(0

.071

8)

(0.0

202)

(0

.027

5)

(0.1

26)

(0.1

07)

(0.0

230)

O

ffsh

orin

g to

oth

er d

evel

oped

cou

ntrie

s 0.

0235

0.

0504

-0

.067

0 0.

117*

* 0.

187

0.05

38

0.00

388

0.01

66

-0.0

270

-8.1

3e-0

5 -0

.054

7

(0.1

44)

(0.1

31)

(0.1

28)

(0.0

544)

(0

.167

) (0

.114

) (0

.018

1)

(0.0

313)

(0

.230

) (0

.200

) (0

.036

2)

N

1096

10

96

1096

10

96

1096

10

96

1096

10

96

1096

10

96

1096

Not

es: T

he m

easu

res

for o

ccup

atio

nal r

estru

ctur

ing

are

for t

he p

erio

d 20

00-2

006,

and

the

outs

ourc

ing

mea

sure

s ar

e fo

r the

per

iod

2001

-200

6.

The

base

line

cate

gory

in a

ll sp

ecifi

catio

ns is

that

the

firm

has

nei

ther

out

sour

ced

dom

estic

ally

nor

off

shor

ed it

s cor

e bu

sine

ss fu

nctio

ns o

ver t

he

perio

d 20

01-2

006.

The

firm

-leve

l m

odel

s ar

e es

timat

ed w

ith e

mpl

oym

ent

wei

ghte

d O

LS, a

s ex

plai

ned

in t

he t

ext.

The

unre

porte

d co

ntro

ls

Tabl

e 1

The

effe

ct o

f out

sour

cing

on

occu

pati

onal

rest

ruct

urin

g at

the

firm

leve

l

Not

es: T

he m

easu

res

for o

ccup

atio

nal r

estr

uctu

ring

are

for t

he p

erio

d 20

00–2

006,

and

the

outs

ourc

ing

mea

sure

s ar

e fo

r the

per

iod

2001

–200

6. T

he b

asel

ine

cate

gory

in a

ll sp

ecifi

catio

ns

is th

at th

e fir

m h

as n

eith

er o

utso

urce

d do

mes

tical

ly n

or o

ffsho

red

its c

ore

busi

ness

func

tions

ove

r the

per

iod

2001

–200

6. T

he fi

rm-le

vel m

odel

s ar

e es

timat

ed w

ith e

mpl

oym

ent w

eigh

ted

OLS

, as

expl

aine

d in

the

text

. The

unr

epor

ted

cont

rols

incl

ude

the

loga

rithm

of e

mpl

oym

ent i

n th

e fir

m a

nd a

set

of i

ndic

ator

s fo

r 22

indu

strie

s. R

obus

t sta

ndar

d er

rors

in p

aren

thes

es: *

**

p<0.

01, *

* p<

0.05

, * p

<0.1

.

14

(PanelB,Columns11).Thisresultimpliesthattheadverseeffectsare,asexpected,concen-tratedonblue-collartasks.Itisparticularlyinterestingtoobservethatoffshoringtootherde-velopedcountriesdoesnothavesignificanteffectsontaskcreationordestructionoronthechangeinthestructureoftasks.However,itdoeshaveastrongpositiveeffectoninternalmo-bility(measuredbyinternaltaskseparation),whichlikelyreflectstheongoingchangesinor-ganizationalstructure(PanelB,Column4).

Theeffectsofdomesticoutsourcingarenotablydifferentfromoffshoring.Interestingly,thisoutsourcinghasanegativeeffectontheshareof interactive tasks in the firms(PanelsA–B,Column7).Theeffectontheshareofnon-routinetasksisalsonegative,butstatisticallyinsig-nificant.Domesticoutsourcingispositivelyassociatedwithinternalseparation(andthusin-ternalhiring),indicatingthatitpromotesintra-firmoccupationalmobility(PanelsA–B,Col-umn4).

Overall,theevidenceclearlyindicatesthatoutsourcinghasanimportanteffectonoccupation-alrestructuringandthatitcanbeexpectedtohavedirectadverseeffectsonthewell-beingofemployeesduetoitseffectofincreasingtaskdestructionandtaskseparationinfirms.Howev-er,theresultsalsorevealthatpartofoccupationalrestructuringtakesplacethroughinternaltaskseparation(andthusinternaltaskhiring).Thistypeofmobilityisnotnecessarilysolelynegativefromthepointofviewoftheemployeesinvolved.

Outsourcing and the Well-being of Employees. Next, we examine how firms’ outsourcing af-fectsdifferentdimensionsofwell-beingamongsurvivingemployees,basedonequation2(Ta-ble2).12Notethatthedescriptivestatisticsfortheoutsourcingvariablesconfirmthattheem-ployee-leveldataremainrepresentativecomparedtotheemployment-weightedfirm-levelda-ta(TableA2,Columns3and5).Beforeexaminingtheeffectsofoutsourcing,wefirstnotethattheoccupationgrouphasasignificantimpactonemployeewell-being.Table2showsthatper-ceivedwell-being isparticularly lowamongserviceand salesworkersaswell as among thetypical blue-collar occupations. The latter group of occupations also has particularly poor(physical)workingconditions(PanelA,Columns3–6).

An important finding is that offshoring does not have an independent, statistically signifi-cant effect on any measure of employee well-being, conditional on occupational group andindividual-levelcontrolvariables(PanelA).Thereis,however,someevidencethatdomesticoutsourcinghasapositiveimpactonjobsatisfactionthatcanbeattributedtoalowerlevelofharm(PanelA,Columns1and3).Theestimatesforthe(unreported)controlvariablesthatareincludedinallspecificationsofTable2areinaccordancewithpreviousstudiesthathaveusedvariousFinnishdatasetstoestimatewell-beingequations.13

ItisnotablethattheestimatesinPanelAofTable2donotrevealasignificantimpactofout-sourcingonperceiveduncertaintyintheFinnishcontext.Previousevidencehasrelatedoff-shoringandothermeasuresofglobalizationtojob-lossfears(e.g.,ScheveandSlaughter2004;Geisheckeretal.2011;Lurweg2010).ThereisalsosomeevidencefortheseeffectsfromotherNordiccountries.Munch(2011)reportsthatoutsourcingsomewhatincreasestheunemploy-mentriskoflow-skilledworkersintheDanishmanufacturingsector.Thesefindingsarerel-

12 The correlations between the variables that capture working conditions are reported in Table A3.13 The estimation results for the control variables are available upon request.

15Outsourcing, Occupational Restructuring, and Employee Well-being: Is There a Silver Lining?

Tabl

e 2

The

effe

ct o

f out

sour

cing

on

perc

eive

d w

ell-b

eing

for a

ll em

ploy

ees

Not

es: T

he v

aria

bles

that

cap

ture

em

ploy

ee w

ell-b

eing

are

mea

sure

d in

200

8. T

he o

utso

urci

ng m

easu

res a

re fo

r the

per

iod

2001

–200

6. T

he e

stim

atio

n sa

mpl

e co

nsis

ts o

f em

ploy

ees w

ho w

ere

empl

oyed

in th

e sa

me

firm

ove

r the

per

iod

2006

–200

8. T

he b

asel

ine

cate

gory

in a

ll sp

ecifi

catio

ns is

that

the

firm

has

nei

ther

out

sour

ced

dom

estic

ally

nor

offs

hore

d its

cor

e bu

sine

ss fu

nctio

ns o

ver t

he p

erio

d 20

01–2

006.

All

spec

ifica

tions

in

clud

e th

e co

ntro

l var

iabl

es th

at a

re li

sted

in T

able

A1.

The

mod

els

also

incl

ude

indi

cato

rs fo

r the

sec

tors

of t

he e

cono

my.

The

bas

elin

e fo

r occ

upat

iona

l gro

ups

cons

ists

of m

anag

ers

and

prof

essi

onal

s. M

argi

nal

effe

cts

from

pro

bit m

odel

s ar

e re

port

ed in

Col

umns

2–8

. Sta

ndar

d er

rors

are

adj

uste

d fo

r clu

ster

ing

at th

e le

vel o

f firm

s fo

r whi

ch e

mpl

oyee

s w

ork.

Sta

tistic

al s

igni

fican

ce: *

** p

<0.0

1, *

* p<

0.05

, * p

<0.1

.

39

TAB

LE 2

THE

EFFE

CT

OF

OU

TSO

UR

CIN

G O

N P

ERC

EIV

ED W

ELL-

BEI

NG

FO

R A

LL E

MPL

OY

EES

Pane

l A

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

Jo

b sa

tisfa

ctio

n U

ncer

tain

ty

Har

m

Haz

ard

No

prom

otio

n N

o vo

ice

Dis

crim

inat

ion

Wor

k in

tens

ity

D

omes

tic o

utso

urci

ng

0.09

05*

-0.0

0043

0 -0

.072

6*

0.00

521

-0.0

144

-0.0

268

-0.0

0757

0.

0195

(0.0

542)

(0

.045

9)

(0.0

371)

(0

.045

4)

(0.0

512)

(0

.043

3)

(0.0

442)

(0

.033

2)

Off

shor

ing

-0.0

512

-0.0

222

-0.0

278

-0.0

536

0.01

26

0.01

96

-0.0

171

0.01

36

(0

.068

3)

(0.0

534)

(0

.049

6)

(0.0

549)

(0

.061

1)

(0.0

471)

(0

.054

3)

(0.0

393)

Te

chni

cian

s and

ass

ocia

te p

rofe

ssio

nals

-0

.057

1 0.

0089

2 -0

.048

5 0.

0770

0.

0123

0.

211*

**

0.01

75

0.18

9***

(0.0

748)

(0

.049

6)

(0.0

557)

(0

.062

0)

(0.0

584)

(0

.033

3)

(0.0

549)

(0

.055

2)

Cle

rical

supp

ort w

orke

rs

-0.1

60

-0.0

607

0.15

0**

0.05

93

0.04

64

0.27

2***

-0

.013

2 0.

143*

(0.1

05)

(0.0

767)

(0

.071

3)

(0.0

936)

(0

.082

6)

(0.0

247)

(0

.073

6)

(0.0

769)

Se

rvic

e an

d sa

le w

orke

rs

-0.2

89**

-0

.122

0.

157

0.26

6***

0.

145

0.22

7***

-0

.045

9 0.

187*

(0.1

31)

(0.1

08)

(0.1

06)

(0.0

902)

(0

.105

) (0

.034

1)

(0.1

05)

(0.0

964)

C

raft

and

rela

ted

trade

wor

kers

-0

.177

* -0

.015

6 0.

158*

0.

231*

**

0.21

8**

0.25

9***

0.

0566

0.

0700

(0.1

04)

(0.0

699)

(0

.084

9)

(0.0

809)

(0

.084

9)

(0.0

300)

(0

.081

1)

(0.0

768)

Pl

ant a

nd m

achi

nery

ope

rato

rs

-0.0

983

0.01

97

0.28

2***

0.

355*

**

0.19

9***

0.

321*

**

-0.0

523

0.16

1**

(0

.095

6)

(0.0

634)

(0

.077

3)

(0.0

677)

(0

.076

4)

(0.0

263)

(0

.067

6)

(0.0

776)

O

ther

wor

kers

-0

.214

0.

0453

0.

200*

* 0.

335*

**

0.29

7***

0.

224*

**

0.06

70

0.25

3***

(0.1

43)

(0.0

719)

(0

.097

9)

(0.0

873)

(0

.088

2)

(0.0

324)

(0

.084

7)

(0.0

943)

N

770

770

770

770

770

770

770

770

Pa

nel B

(1

) (2

) (3

) (4

) (5

) (6

) (7

) (8

)

Job

satis

fact

ion

Unc

erta

inty

H

arm

H

azar

d N

o pr

omot

ion

No

voic

e D

iscr

imin

atio

n W

ork

inte

nsity

Dom

estic

out

sour

cing

0.

102*

-0

.006

58

.. ..

0.00

752

-0.0

0908

0.

0014

3 0.

0258

(0.0

546)

(0

.046

4)

(0.0

523)

(0

.043

8)

(0.0

452)

(0

.034

2)

Off

shor

ing

to th

e 15

EU

cou

ntrie

s 0.

104

0.01

91

0.00

759

-0.0

145

-0.1

18

-0.0

359

(0

.104

) (0

.086

5)

(0.1

00)

(0.0

862)

(0

.086

3)

(0.0

493)

O

ffsh

orin

g to

the

rest

of E

urop

e 0.

0133

-0

.154

*

0.

0852

0.

0738

0.

0699

0.

150*

*

(0.0

979)

(0

.081

7)

(0.0

990)

(0

.057

4)

(0.0

751)

(0

.064

0)

Off

shor

ing

to d

evel

opin

g co

untri

es

-0.2

06*

0.06

72

-0.1

05

-0.1

14

-0.0

746

-0.0

844*

*

(0.1

06)

(0.0

747)

(0

.076

2)

(0.1

01)

(0.0

831)

(0

.034

6)

Off

shor

ing

to o

ther

dev

elop

ed c

ount

ries

0.09

91

0.08

81

-0.2

36**

-0

.089

9 -0

.099

6 -0

.015

9

(0.0

921)

(0

.090

8)

(0.1

07)

(0.1

37)

(0.1

19)

(0.0

860)

Te

chni

cian

s and

ass

ocia

te p

rofe

ssio

nals

-0

.056

9 0.

0087

9

0.

0117

0.

210*

**

0.01

66

0.18

7***

(0.0

748)

(0

.049

5)

(0.0

588)

(0

.033

5)

(0.0

553)

(0

.054

9)

Cle

rical

supp

ort w

orke

rs

-0.1

70*

-0.0

575

0.03

89

0.26

9***

-0

.016

9 0.

132*

(0.1

03)

(0.0

770)

(0

.082

7)

(0.0

250)

(0

.072

1)

(0.0

748)

40

Serv

ice

and

sale

wor

kers

-0

.289

**

-0.1

23

0.14

3 0.

226*

**

-0.0

491

0.18

4**

(0

.133

) (0

.107

)

(0

.106

) (0

.034

2)

(0.1

03)

(0.0

902)

C

raft

and

rela

ted

trade

wor

kers

-0

.170

-0

.011

1

0.

213*

* 0.

257*

**

0.05

13

0.06

56

(0

.104

) (0

.069

9)

(0.0

853)

(0

.030

3)

(0.0

811)

(0

.075

6)

Plan

t and

mac

hine

ry o

pera

tors

-0

.096

6 0.

0215

0.

198*

**

0.32

2***

-0

.052

2 0.

155*

*

(0.0

956)

(0

.062

9)

(0.0

769)

(0

.026

1)

(0.0

679)

(0

.076

8)

Oth

er w

orke

rs

-0.2

18

0.04

25

0.29

3***

0.

222*

**

0.06

62

0.25

0***

(0.1

43)

(0.0

724)

(0

.089

1)

(0.0

327)

(0

.085

0)

(0.0

932)

N

770

770

770

770

770

770

Not

es: T

he v

aria

bles

that

cap

ture

em

ploy

ee w

ell-b

eing

are

mea

sure

d in

200

8. T

he o

utso

urci

ng m

easu

res

are

for

the

perio

d 20

01-2

006.

The

es

timat

ion

sam

ple

cons

ists

of

empl

oyee

s w

ho w

ere

empl

oyed

in

the

sam

e fir

m o

ver

the

perio

d 20

06-2

008.

The

bas

elin

e ca

tego

ry i

n al

l sp

ecifi

catio

ns i

s th

at t

he f

irm h

as n

eith

er o

utso

urce

d do

mes

tical

ly n

or o

ffsh

ored

its

core

bus

ines

s fu

nctio

ns o

ver

the

perio

d 20

01-2

006.

All

spec

ifica

tions

incl

ude

the

cont

rol v

aria

bles

that

are

list

ed in

Tab

le A

1. T

he m

odel

s al

so in

clud

e in

dica

tors

for t

he s

ecto

rs o

f the

eco

nom

y. T

he

base

line

for

occu

patio

nal g

roup

s co

nsis

ts o

f m

anag

ers

and

prof

essi

onal

s. M

argi

nal e

ffec

ts f

rom

pro

bit m

odel

s ar

e re

porte

d in

Col

umns

2-8

. St

anda

rd e

rror

s ar

e ad

just

ed f

or c

lust

erin

g at

the

leve

l of

firm

s fo

r w

hich

em

ploy

ees

wor

k. S

tatis

tical

sig

nific

ance

: ***

p<0

.01,

**

p<0.

05, *

p<

0.1

16

evanttothesurvivingemployeesbecausepreviousoutsourcingmayincreasetheunemploy-mentriskforyearstocome.

ThespecificationsdocumentedinPanelAofTable2assumethatthepotentialeffectsofout-sourcingonemployeewell-beingareuniform,irrespectiveofthegeographicaldestinationofoutsourcing.Adetailedbreakdownofoffshoringbygeographicaldestinationrevealsinterest-ingheterogeneityintheeffects(Table2,PanelB).14Offshoringtodevelopingcountrieshavesomenegativeeffectonoverallmeasureofemployeewell-being(i.e.,jobsatisfaction)(PanelB,Column1).

AnotherinterestingfindingisthatoffshoringtotheUS,Canada,andotherdevelopedcoun-tries significantlyraises survivingemployees’perceptionsofpromotionandwageprospects(Table2,PanelB,Column5).ThisobservationisreasonablebecauseFinlandisasmall,openeconomywith limitedopportunities,especially forhighlyskilledworkers.Thus,offshoring,alongwithotheraspectsofglobalization,createsopportunitiestoadvanceone’scareerbecauseiteffectivelybroadensthemarketfortalent.Offshoringamongdevelopedcountries ismostlikelytobereciprocal,andworkersinhigh-wagecountriesarecomplementsratherthansub-stitutes (Geisheckeretal.2011).Thequantitativemagnitudeof thiseffect is substantialbe-causeoffshoringtotheUS,Canada,andotherdevelopedcountriesdecreasesemployees’per-ceptions of poor promotion prospects by 24%. This result is consistent with the pattern inTable1(PanelB,Column4),accordingtowhichoffshoringtootherdevelopedcountriessub-stantiallyincreasesinternaltaskseparation,whichmeasurestheverticalmobilityofemploy-ees infirms.Therefore, thevertical fragmentationofproductioniscloselyconnectedtotheverticalmobilityofemployeesinfirmsthatoffshore.Furthermore,thereisevidenceforan-otherpositiveeffectofoffshoringonemployeewell-being:theresultssuggestthatoffshoringtodevelopingcountriesdecreasesperceivedworkintensityamongsurvivingemployees(Ta-ble2,PanelB,Column8).However,offshoringtotherestofEuropeincreasesworkintensity.

Table3documentsa separate setofestimates forknowledgeworkersbecause thisgroupofemployees is more likely to benefit from outsourcing in terms of well-being than are otherworkergroups.TheresultsinPanelArevealthatoffshoringdecreasestheprevalenceofharm (Column3).Therearealsosignificanteffectsregardingdiscrimination (Column7).However,theseeffectsarecontradictoryinthatdomesticoutsourcingincreasestheperceptionofdis-criminationamongsurvivingknowledgeworkers,butoffshoringseemstodecreaseit.Thees-timates that break down according to the geographical destination of offshoring show thatoffshoring to the15EUcountriesdecreases theperceptionofdiscrimination. It is alsono-tablethatdomesticoutsourcing,alongwithoffshoringtothe15EUcountries,improvesthelevelofjobsatisfaction(PanelB,Column1).However,negativewell-beingeffectsforknowl-edgeworkersalsoexistbecauseoffshoringtodevelopingcountriessignificantlyincreasestheperceptionofuncertainty(PanelB,Column2).Thepatterninwhichoffshoringtootherde-velopedcountriesconsiderablyimprovespromotionprospectsremainsintactforknowledgeworkers(PanelB,Column5).

Occupational Restructuring and the Well-being of Employees. The concluding section of themainestimationresultsexplorestheeffectofoccupationalrestructuringamongthecontinu-

14 We do not report the results for harm and hazard for these specifications because some cells of the data contain too few observa-tions to obtain reliable estimates.

17Outsourcing, Occupational Restructuring, and Employee Well-being: Is There a Silver Lining?

41

TAB

LE 3

THE

EFFE

CT

OF

OU

TSO

UR

CIN

G O

N P

ERC

EIV

ED W

ELL-

BEI

NG

FO

R K

NO

WLE

DG

E W

OR

KER

S

Pane

l A

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

Jo

b sa

tisfa

ctio

n U

ncer

tain

ty

Har

m

Haz

ard

No

prom

otio

n N

o vo

ice

Dis

crim

inat

ion

Wor

k in

tens

ity

D

omes

tic o

utso

urci

ng

0.12

9 0.

0308

-0

.041

1 0.

0377

0.

0579

-0

.008

31

0.09

45*

0.05

01

(0

.080

6)

(0.0

540)

(0

.035

6)

(0.0

575)

(0

.068

0)

(0.0

623)

(0

.053

2)

(0.0

449)

O

ffsh

orin

g -0

.016

6 -0

.089

9 -0

.085

4**

-0.0

655

-0.0

487

0.03

25

-0.1

47**

-0

.043

0

(0.1

04)

(0.0

797)

(0

.038

0)

(0.0

625)

(0

.074

9)

(0.0

730)

(0

.068

2)

(0.0

404)

N

421

421

421

421

421

421

421

421

Pa

nel B

(1

) (2

) (3

) (4

) (5

) (6

) (7

) (8

)

Job

satis

fact

ion

Unc

erta

inty

H

arm

H

azar

d N

o pr

omot

ion

No

voic

e D

iscr

imin

atio

n W

ork

inte

nsity

Dom

estic

out

sour

cing

0.

152*

0.

0133

..

.. 0.

0968

0.

0164

0.

104*

* 0.

0592

(0.0

796)

(0

.053

5)

(0.0

721)

(0

.064

2)

(0.0

525)

(0

.045

1)

Off

shor

ing

to th

e 15

EU

cou

ntrie

s 0.

230*

-0

.073

8

0.

0896

0.

0435

-0

.180

* -0

.014

8

(0.1

22)

(0.1

21)

(0.1

29)

(0.1

15)

(0.1

05)

(0.0

640)

O

ffsh

orin

g to

the

rest

of E

urop

e 0.

0297

-0

.231

*

-0

.058

1 0.

0872

0.

0048

2 0.

0843

(0.1

57)

(0.1

20)

(0.1

02)

(0.0

939)

(0

.106

) (0

.070

2)

Off

shor

ing

to d

evel

opin

g co

untri

es

-0.2

31

0.04

26

-0.1

23

-0.0

887

-0.1

81

-0.1

03**

*

(0.1

78)

(0.1

06)

(0.1

15)

(0.1

34)

(0.1

17)

(0.0

239)

O

ffsh

orin

g to

oth

er d

evel

oped

cou

ntrie

s -0

.008

15

0.16

9**

-0.1

75**

-0

.180

-0

.094

1 -0

.029

2

(0.0

921)

(0

.090

8)

(0.0

835)

(0

.163

) (0

.129

) (0

.084

8)

N

42

1 42

1

42

1 42

1 42

1 42

1

Not

es: T

he v

aria

bles

that

cap

ture

em

ploy

ee w

ell-b

eing

are

mea

sure

d in

200

8. T

he o

utso

urci

ng m

easu

res

are

for

the

perio

d 20

01-2

006.

The

es

timat

ion

sam

ple

cons

ists

of

empl

oyee

s w

ho w

ere

empl

oyed

in

the

sam

e fir

m o

ver

the

perio

d 20

06-2

008.

The

bas

elin

e ca

tego

ry i

n al

l sp

ecifi

catio

ns i

s th

at t

he f

irm h

as n

eith

er o

utso

urce

d do

mes

tical

ly n

or o

ffsh

ored

its

core

bus

ines

s fu

nctio

ns o

ver

the

perio

d 20

01-2

006.

All

spec

ifica

tions

inc

lude

the

con

trol

varia

bles

that

are

lis

ted

in T

able

A1.

The

mod

els

also

inc

lude

ind

icat

ors

for

the

sect

ors

of t

he e

cono

my.

M

argi

nal e

ffec

ts f

rom

pro

bit m

odel

s ar

e re

porte

d in

Col

umns

2-8

. Sta

ndar

d er

rors

are

adj

uste

d fo

r cl

uste

ring

at th

e le

vel o

f fir

ms

for

whi

ch

empl

oyee

s wor

k. S

tatis

tical

sign

ifica

nce:

***

p<0

.01,

**

p<0.

05, *

p<0

.1

Tabl

e 3

The

effe

ct o

f out

sour

cing

on

perc

eive

d w

ell-b

eing

for k

now

ledg

e w

orke

rs

Not

es: T

he v

aria

bles

that

cap

ture

em

ploy

ee w

ell-b

eing

are

mea

sure

d in

200

8. T

he o

utso

urci

ng m

easu

res

are

for t

he p

erio

d 20

01–2

006.

The

est

imat

ion

sam

ple

cons

ists

of e

mpl

oyee

s w

ho

wer

e em

ploy

ed in

the

sam

e fir

m o

ver t

he p

erio

d 20

06–2

008.

The

bas

elin

e ca

tego

ry in

all

spec

ifica

tions

is t

hat

the

firm

has

nei

ther

out

sour

ced

dom

estic

ally

nor

offs

hore

d its

cor

e bu

si-

ness

func

tions

ove

r the

per

iod

2001

–200

6. A

ll sp

ecifi

catio

ns in

clud

e th

e co

ntro

l var

iabl

es t

hat

are

liste

d in

Tab

le A

1. T

he m

odel

s al

so in

clud

e in

dica

tors

for t

he s

ecto

rs o

f the

eco

nom

y.

Mar

gina

l effe

cts

from

pro

bit

mod

els

are

repo

rted

in C

olum

ns 2

–8. S

tand

ard

erro

rs a

re a

djus

ted

for

clus

terin

g at

the

leve

l of

firm

s fo

r w

hich

em

ploy

ees

wor

k. S

tatis

tical

sig

nific

ance

: **

* p<

0.01

, **

p<0.

05, *

p<0

.1.

18

ingfirmsbetween2000and2006onsurvivingemployees’well-beingtwoyearslater,in2008(Table4).Weuseexactlythesamemeasuresofperceivedwell-beingasthedependentvaria-blesusedinTables2–3.Themostimportantfindingofthespecifications,basedonequation(3),isthatthereisgenerallynostatisticallysignificantrelationshipbetweenoccupationalre-structuringandsubsequentemployeewell-being(Table4).Thus, itseemsthatoccupationalrestructuringdoesnothavesignificantadditionalnegativeeffectsonemployeesbeyonditsdi-rectadverseeffects,documentedinTable1.ThisgeneralpatternisconsistentwiththeresultsinBöckermanetal.(2011),whohavereportedthataverageuncertaintyattheestablishmentleveldoesnotcausesignificantlossesinworksatisfactionintheFinnishcontext.

Occupational restructuring,measuredbychanges in the sharesof interactiveandnon-rou-tinetasks,significantlyreducestheexperienceoflackingavoiceintheworkplace,especiallyamongknowledgeworkers(Table4,PanelB,Column6).ThisresultisinterestingbecauseAu-toretal.(2003)stresstheimportanceofskillcontentinjobtasks.Furthermore,thereisevi-dencethatanincreaseintheshareofknowledgeworkersinafirmsignificantlyimprovesem-ployees’experienceofhavingavoiceintheworkplace(Table4,PanelB,Column6)and,con-sistentwiththatoutcome,increasesjobsatisfaction(Table4,PanelB,Column1).

Additional Aspects. Toshedmorelightontheeffectsofoffshoring,weestimatedasetofaddi-tionalspecifications.Webrieflydiscusstheseresultswithoutpresentingthemintables.Asarobustnesscheck,weperformedtheestimationsofTable1usingunweightedregressions,withandwithoutasizerestrictionforfirms(i.e.,theinclusionoffirmsemployingfewerthan100employees).ThebaselineresultsinTable1seemtoremainlargelyintactinunweightedregres-sionsandinthosewithoutasizerestriction.However,therearealsosomeimportantdiffer-ences;intheseresults,offshoringisnotnegativelyassociatedwithtaskcreationandtaskhir-ing.Furthermore,offshoringdoesnothaveapositiveimpactontheshareofknowledgework-ersinthefirms.Thus,itisimportanttouseemployment-weightedregressionstoempiricallycapturethecreativesideofoffshoring.Furthermore,theresultsinTable1(PanelB,Column1),whichsuggest thatoffshoringtodevelopingcountries leads to taskdestructionandtaskseparation,donotprevailinunweightedregressions.

OneofthestrengthsoftheISSforoutsourcingactivitiesisthatitcontainsinformationaboutfirms’self-declaredmotivationsforconductingoffshoring.Thereisevidencethattheeffectsonemployeewell-beingdiffersignificantlyaccordingtothemotivationforoffshoring.Inpar-ticular,animportantresultinTable2(PanelB,Column5),whichrevealsthatoffshoringsig-nificantly improves surviving employees’ perception of promotion prospects, prevails onlywhenoffshoringismotivatedbyopeningnewmarketsforproductsandservicesratherthanbyeffortstoreducelaborcosts.(Furthermore,thereisasignificantdecreaseintheperceptionofdiscrimination.)Thisfindingislogicalbecausethistypeofoffshoringconstitutessubstan-tialopportunitiesforcareeradvancement,especiallyforknowledgeworkers.

Finally,wefind that insourcing(i.e., theoppositeofoffshoring)generally improvesemployeewell-being. In particular, the results suggest that insourcing significantly decreases the likeli-hoodofperceivedharmsandhazardsintheworkplace.ThisfindingisreasonablebecausethecoststructureishigherinFinlandthaninseveralothercountriesthatwerepreviouslylocationsfortheseactivities.ThetypesofjobsthatareinsourcedtoFinlandarehigh-qualityjobswithhighwagesandamenitiesthatsupporttheperceptionofgoodworkingconditionsamongtheaffect-edemployees.Thereisalsosomeevidencethatinsourcingsupportsoverallsatisfactionatwork.

19Outsourcing, Occupational Restructuring, and Employee Well-being: Is There a Silver Lining?

42

TAB

LE 4

THE

EFFE

CT

OF

OC

CU

PATI

ON

AL

RES

TRU

CTU

RIN

G O

N P

ERC

EIV

ED W

ELL-

BEI

NG

Pane

l A: A

ll em

ploy

ees

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

Jo

b sa

tisfa

ctio

n U

ncer

tain

ty

Har

m

Haz

ard

No

prom

otio

n N

o vo

ice

Dis

crim

inat

ion

Wor

k in

tens

ity

Ta

sk d

estru

ctio

n -0

.030

5 0.

0311

-0

.008

53

-0.0

0628

0.

0033

9 -0

.039

3 0.

0035

1 0.

0241

Ta

sk se

para

tion

-0.0

578

0.04

15

0.01

11

-0.0

201

-0.0

347

-0.0

211

0.05

22

0.04

06

Exte

rnal

task

sepa

ratio

n -0

.061

1 0.

0518

0.

0064

0 -0

.037

8 -0

.014

8 -0

.007

28

0.00

871

-0.0

0627

In

tern

al ta

sk se

para

tion

(i.e.

, int

erna

l ta

sk h

iring

) -0

.032

3 -0

.010

6 0.

0268

0.

0512

-0

.103

-0

.072

7 0.

216*

0.

194*

**

Exce

ss ta

sk re

allo

catio

n -0

.107

0.

0310

-0

.020

1 0.

0330

0.

0244

-0

.031

9 0.

0207

0.

0286

W

orke

r chu

rnin

g -0

.046

9 0.

0226

0.

0259

-0

.020

8 -0

.053

6 0.

0159

0.

0717

0.

0303

Th

e ch

ange

in th

e sh

are

of in

tera

ctiv

e ta

sks

-0.0

772

-0.6

03**

-0

.032

4 0.

332

-0.0

773

-0.4

60*

-0.0

422

-0.1

42

The

chan

ge in

the

shar

e of

non

-rou

tine

task

s 0.

121

-0.3

95**

0.

0956

0.

288

-0.0

578

-0.2

58

-0.0

0415

-0

.016

4

Task

cre

atio

n -0

.010

8 0.

0389

0.

0015

3 0.

0247

-0

.015

3 -0

.000

434

0.00

171

-0.0

118

Task

hiri

ng

-0.0

262

0.05

61

0.00

890

0.02

64

-0.0

338

0.00

352

0.02

07

-0.0

0750

Th

e ch

ange

in th

e sh

are

of k

now

ledg

e w

orke

rs

0.17

0 -0

.106

-0

.018

7 -0

.037

0 0.

0008

36

-0.2

12**

-0

.027

3 0.

0200

Pa

nel B

: Kno

wle

dge

wor

kers

(1

) (2

) (3

) (4

) (5

) (6

) (7

) (8

)

Job

satis

fact

ion

Unc

erta

inty

H

arm

H

azar

d N

o pr

omot

ion

No

voic

e D

iscr

imin

atio

n W

ork

inte

nsity

Task

des

truct

ion

0.03

96

-0.0

0084

5 0.

0342

0.

0022

6 -0

.043

9 -0

.082

1 -0

.069

6 -0

.043

2 Ta

sk se

para

tion

-0.0

521

0.02

82

0.06

91

0.03

19

-0.0

0474

-0

.044

6 0.

0459

-0

.009

26

Exte

rnal

task

sepa

ratio

n -0

.050

5 0.

0301

0.

0762

0.

0072

4 -0

.002

56

-0.0

382

-0.0

0134

-0

.060

7 In

tern

al ta

sk se

para

tion

(i.e.

, int

erna

l ta

sk h

iring

) -0

.054

3 0.

0180

0.

0298

0.

114

-0.0

117

-0.0

653

0.21

2 0.

128

Exce

ss ta

sk re

allo

catio

n -0

.044

5 0.

0462

0.

0399

0.

0423

0.

0947

0.

0052

1 -0

.010

9 0.

0392

W

orke

r chu

rnin

g -0

.131

0.

0458

0.

0650

0.

0458

0.

0458

0.

0328

0.

153*

* 0.

0287

Th

e ch

ange

in th

e sh

are

of in

tera

ctiv

e ta

sks

-0.1

51

-0.5

55*

0.14

6 0.

181

-0.0

630

-1.1

81**

* 0.

0087

4 -0

.289

The

chan

ge in

the

shar

e of

non

-rou

tine

task

s 0.

250

-0.3

86*

0.12

8 0.

223

-0.2

02

-0.8

03**

* -0

.136

-0

.089

6

Task

cre

atio

n -0

.019

2 0.

0424

-0

.000

484

0.03

72

0.03

58

0.04

50

0.00

444

0.01

85

Task

hiri

ng

-0.0

564

0.06

69

0.01

55

0.06

01

0.05

80

0.06

70

0.04

15

0.03

15

The

chan

ge in

the

shar

e of

kno

wle

dge

wor

kers

0.

360*

-0

.175

-0

.010

3 -0

.042

0 -0

.075

4 -0

.459

***

-0.1

33

-0.0

302

Tabl

e 4

The

effe

ct o

f occ

upat

iona

l res

truc

turi

ng o

n pe

rcei

ved

wel

l-bei

ng

Not

es: T

he v

aria

bles

that

cap

ture

em

ploy

ee w

ell-b

eing

are

mea

sure

d in

200

8. T

he m

easu

res

for o

ccup

atio

nal r

estr

uctu

ring

are

for t

he p

erio

d 20

00–2

006

and

are

calc

ulat

ed o

nly

for c

on-

tinuo

us fi

rms.

The

sam

ple

cons

ists

of e

mpl

oyee

s w

ho w

ere

empl

oyed

in th

e sa

me

firm

ove

r the

per

iod

2006

–200

8. E

ach

cell

of th

e ta

ble

repo

rts

the

para

met

er e

stim

ate

from

a s

epar

ate

spec

ifica

tion.

All

mod

els

incl

ude

the

cont

rol v

aria

bles

that

are

list

ed in

Tab

le A

1. M

argi

nal e

ffect

s fr

om p

robi

t mod

els

are

repo

rted

in C

olum

ns 2

–8. T

he s

ampl

e si

ze is

1,1

74 in

all

spec

ifica

-tio

ns. T

he u

nrep

orte

d st

anda

rd e

rror

s ar

e ad

just

ed fo

r clu

ster

ing

at th

e le

vel o

f firm

s fo

r whi

ch e

mpl

oyee

s w

ork.

Sta

tistic

al s

igni

fican

ce: *

** p

<0.0

1, *

* p<

0.05

, * p

<0.1

.

20

6 ConclusionsThispaperexaminestheeffectsofoutsourcingonvariousmeasuresofemployeewell-beingintheFinnishcontext.Thewell-beinglossescausedbytheverticalfragmentationofproductionpotentiallyconstituteanimportantpartoftheshort-runadjustmentcoststoemployees,andtheycanexplainthepersistentresistancetooutsourcing.Weuseparticularlyrichmatchedda-tathatcombineafirm-levelsurveyofoutsourcingwithasurveyofemployeesthatcontainsdetailedinformationonseveralaspectsofsubjectivewell-being.Furthermore,weanalyzetheconnectionofoutsourcingtocomprehensivemeasuresofoccupationalrestructuring.

Asidefromthedirectnegativeeffectofoffshoringonemployees(attributabletogreater jobdestructionandworkeroutflow),basedontheevidence,therelationshipbetweenoutsourcingandemployeewell-beingiscomplex.Themainfindingofthispaperisthattheresultsstrong-lypointtothesubstantialheterogeneityintheeffectsofoffshoring.Forexample,wefindthattherelationshipbetweenoutsourcingandperceivedwell-beingdiffersconsiderablyaccordingtothegeographicaldestinationofoffshoring.Thereissomeevidencethatoffshoringtodevel-opingcountriesreducesoverallsatisfactionatwork.Oneexplanationforthisoutcomeisthatoffshoringtolow-wagecountriessubstitutesfordomesticemployment(HarrisonandMcMil-lan2011).

Aside from these negative effects, offshoring also has plausible positive effects on employ-eewell-beingthathavelargelybeenoverlookedintheexistingliterature.Wedocumentthathigheroccupationalstatusclearlyimproveswell-beingatwork,andoffshoringconsiderablyincreasestheshareofknowledgeworkersinfirms.Importantly,inadditiontothisdirectcom-positioneffect,thereisalsoevidencefortheexistenceofpositiveindependenteffects.There-fore,evengiventheprevailingstructureofdifferentoccupationsinthefirms,anincreaseintheshareofknowledgeworkersimprovessomeimportantaspectsofperceivedemployeewell-being.

Furthermore, we find that offshoring to other developed countries improves surviving em-ployees’promotionandsubsequentwageprospects.Theseeffectsareparticularlypronouncedforknowledgeworkersandwhenoffshoringhasbeenmotivatedbyopeningnewmarketsforproductsandservices.TheseeffectsareconsistentwiththefindingsbyHickmanandOlney(2011),whoarguethatemployeeshaverespondedtooffshoringbyincreasingtheirstockofhumancapitalbyacquiringbettereducationintheU.S.context.Bettereducationcreatesop-portunities for careeradvancementbecause themarket for talentedworkers is larger (Kau-hanenandNapari2011).Thefactthatpromotionprospectsareparticularlysensitivetooff-shoring is also reasonable because harms and hazards are closely related to the fixed stockofcapital thatconstitutesthephysicalworkenvironment,whichdoesnotchangerapidly infirms.Incontrast,promotionprospectsarerelatedtoexpectationsthatcanchangerapidlyasafirmchanges.Therefore,thebottomlineofthefindingsisthatoffshoringnotonlyhasnega-tiveeffectsonthewell-beingofsurvivingemployees,butitseffectsdiffersubstantiallybythetypeofoffshoringandbyoutcomes.

Astraightforwardgeneralizationofourresultsatthelevelofthewholeeconomyinvolvesapo-tentialfallacyofcomposition.Theestimatedpositivewell-beingeffectsonsurvivingemploy-eesdonotprovideacompletepictureofthetotalimpactofoffshoringonwell-beingbecause

21Outsourcing, Occupational Restructuring, and Employee Well-being: Is There a Silver Lining?

thosewhoendupunemployedareexcluded.Therefore,empirically,abroaderandmorebal-ancedpicturewouldrequireanalysesofwhathappenstothoseindividualswholosetheirjobsduetooffshoring.Whatisthequalityofjobsforwhichtheyarehired?Theempiricalevidenceis somewhat reassuring.A largemajorityof theunemployedwill findanew jobeventually,andthenewlycreatedjobsareusuallymoreproductivethanthedestroyedoldjobs.Forexam-ple,previousFinnishevidencesuggeststhatasignificantproportionofaggregateproductivitygrowthcanbeattributedtothecreationofnewjobsandthedestructionofoldjobs,atleastinmanufacturing(Malirantaetal.2010).Incontrast,focusingonthepost-recessionyears1992–1997,IlmakunnasandMaliranta(2004)showthatoldandlow-productivityplantshavehighseparation rates to unemployment, and new and high-productivity plants have high hiringratesfromunemployment.Intermsofsocialpolicy,theprimarychallengeistobothstrength-enthepositiveeffectsofoffshoring-triggeredrestructuringandtofacilitateadjustmenttothenegativeimpacts,includinggreaterturbulenceandpolarizationinthelabormarkets.

22

Appendix

44

APPENDIX

TABLE A1

DEFINITIONS OF VARIABLES

Variable Definition/measurement The measures of perceived employee well-being (QWLS)

Job satisfaction Job satisfaction is measured by means of alternatives 1 (very satisfied), 2

(quite satisfied), 3 (rather dissatisfied), and 4 (very dissatisfied). Uncertainty Work including at least one insecurity factor (includes transfer to other duties,

threat of temporary dismissal, threat of permanent dismissal, threat of unemployment, threat of becoming incapable of work, unforeseen changes, work load increasing beyond tolerance) = 1, otherwise = 0.

Harm At least one adverse factor that affects work ‘very much’ (includes heat, cold, vibration, draft, noise, smoke, gas and fumes, humidity, dry indoor air, dust, dirtiness of work environment, poor or glaring lighting, irritating or corrosive substances, restless work environment, repetitive and monotonous movements, difficult or uncomfortable working positions, time pressure and tight time schedules, heavy lifting, lack of space, mildew in buildings) = 1, otherwise = 0.

Hazard At least one factor experienced as ‘a distinct hazard’ (includes accident risk, becoming subject to physical violence, hazards caused by chemical substances, radiation hazard, major catastrophe hazard, hazard of infectious diseases, hazard of skin diseases, cancer risk, risk of strain injuries, risk of succumbing to mental disturbance, risk of grave work exhaustion, risk of causing serious injury to others, risk of causing serious damage to valuable equipment or product) = 1, otherwise = 0.

No promotion Advancement opportunities in current workplace ‘poor’ = 1, otherwise = 0. No voice ‘Not at all’ able to influence at least one factor at work (includes content of

tasks, order in which tasks are completed, pace of work, working methods, division of tasks between employees, choice of working partners, equipment purchases) = 1, otherwise = 0.

Discrimination Experience of at least one type of unequal treatment or discrimination in current workplace (includes time of hiring, remuneration, career advancement opportunities, access to training arranged by employer, receiving information, attitudes of co-workers or superiors) = 1, otherwise = 0.

Work intensity Intensity at work is high enough to cause sickness absence = 1, otherwise 0.

The measures of outsourcing (ISS)

Domestic outsourcing Firm has domestically outsourced its core business functions (i.e.,

production of goods and/or services) over the period 2001-2006 = 1, otherwise 0.

Offshoring (i.e. international outsourcing)

Firm has offshored abroad its core business functions over the period 2001-2006 = 1, otherwise 0.

Offshoring to the 15 EU countries

Firm has offshored its core business functions to the 15 EU countries over the period 2001-2006 = 1, otherwise 0. The EU 15 countries are Belgium, Denmark, Germany, Greece, Spain, France, Ireland, Italy, Luxembourg, Netherlands, Austria, Portugal, Sweden, and the United Kingdom. Finland is excluded from the list of the EU 15 countries.

Offshoring to the rest of Europe Firm has offshored its core business functions to the rest of Europe over the period 2001-2006 = 1, otherwise 0. The rest of Europe includes 12 EU countries (i.e., Czech Republic, Estonia, Cyprus, Latvia, Lithuania, Hungary, Malta, Poland, Slovenia, Slovak Republic, Bulgaria, and Romania) and Switzerland, Norway, Turkey, Russia, Belo Russia, Ukraine, and the Balkan states.

Offshoring to developing countries

Firm has offshored its core business functions to developing countries over the period 2001-2006 = 1, otherwise 0. The developing countries include China, India, South and Central America (including Mexico), and Africa.

Offshoring to other developed countries

Firm has offshored its core business functions to other developed countries over the period 2001-2006 = 1, otherwise 0. The other

Table A1 Definitions of variables

23Outsourcing, Occupational Restructuring, and Employee Well-being: Is There a Silver Lining?

45

developed countries include the US, Canada, Japan, Korea, the countries of the Near East and Far East, and Oceania.

Control variables Human capital (QWLS) Female 1 = female, 0 = male. Age <=34 Age <= 34 = 1, otherwise = 0. Age 35-44 Age 35-44 = 1, otherwise = 0 (reference) Age 45-54 Age 45-54 = 1, otherwise = 0 Age 55-64 Age 55-64 = 1, otherwise = 0 Married Married = 1, otherwise = 0. Basic education only Less than second stage of secondary level education (International

Standard Classification of Education (ISCED) 0-2) = 1, otherwise = 0 (reference).

Middle education Second stage of secondary level education (ISCED 3) = 1, otherwise = 0 Higher education Third-level education (ISCED 5-7) = 1, otherwise = 0 Union member Member of trade union = 1, otherwise = 0. Work history (FLEED) Past earnings A logarithm of past average earnings over the period 1990-2007, deflated

to the year 2000 by using the consumer price index. Past employment The total number of employment months over the period 1990-2007. Past unemployment The total number of unemployment months over the period 1990-2007. Self-assessed health (QWLS) Self-assessment of working capacity. The variable is scaled from 0 (total

inability to work) to 10 (top condition). Employer characteristics (QWLS)

Plant size <100 Size of plant under 100 employees = 1, otherwise = 0 (reference) Plant size 100-249 Size of plant 100-249 employees = 1, otherwise = 0 Plant size 250-999 Size of plant 250-999 employees = 1, otherwise = 0 Plant size > 1000 Size of plant over 1000 employees = 1, otherwise = 0

Note: The measures of occupational restructuring are defined in the text. Note: The measures of occupational restructuring are defined in the text.

44

APPENDIX

TABLE A1

DEFINITIONS OF VARIABLES

Variable Definition/measurement The measures of perceived employee well-being (QWLS)

Job satisfaction Job satisfaction is measured by means of alternatives 1 (very satisfied), 2

(quite satisfied), 3 (rather dissatisfied), and 4 (very dissatisfied). Uncertainty Work including at least one insecurity factor (includes transfer to other duties,

threat of temporary dismissal, threat of permanent dismissal, threat of unemployment, threat of becoming incapable of work, unforeseen changes, work load increasing beyond tolerance) = 1, otherwise = 0.

Harm At least one adverse factor that affects work ‘very much’ (includes heat, cold, vibration, draft, noise, smoke, gas and fumes, humidity, dry indoor air, dust, dirtiness of work environment, poor or glaring lighting, irritating or corrosive substances, restless work environment, repetitive and monotonous movements, difficult or uncomfortable working positions, time pressure and tight time schedules, heavy lifting, lack of space, mildew in buildings) = 1, otherwise = 0.

Hazard At least one factor experienced as ‘a distinct hazard’ (includes accident risk, becoming subject to physical violence, hazards caused by chemical substances, radiation hazard, major catastrophe hazard, hazard of infectious diseases, hazard of skin diseases, cancer risk, risk of strain injuries, risk of succumbing to mental disturbance, risk of grave work exhaustion, risk of causing serious injury to others, risk of causing serious damage to valuable equipment or product) = 1, otherwise = 0.

No promotion Advancement opportunities in current workplace ‘poor’ = 1, otherwise = 0. No voice ‘Not at all’ able to influence at least one factor at work (includes content of

tasks, order in which tasks are completed, pace of work, working methods, division of tasks between employees, choice of working partners, equipment purchases) = 1, otherwise = 0.

Discrimination Experience of at least one type of unequal treatment or discrimination in current workplace (includes time of hiring, remuneration, career advancement opportunities, access to training arranged by employer, receiving information, attitudes of co-workers or superiors) = 1, otherwise = 0.

Work intensity Intensity at work is high enough to cause sickness absence = 1, otherwise 0.

The measures of outsourcing (ISS)

Domestic outsourcing Firm has domestically outsourced its core business functions (i.e.,

production of goods and/or services) over the period 2001-2006 = 1, otherwise 0.

Offshoring (i.e. international outsourcing)

Firm has offshored abroad its core business functions over the period 2001-2006 = 1, otherwise 0.

Offshoring to the 15 EU countries

Firm has offshored its core business functions to the 15 EU countries over the period 2001-2006 = 1, otherwise 0. The EU 15 countries are Belgium, Denmark, Germany, Greece, Spain, France, Ireland, Italy, Luxembourg, Netherlands, Austria, Portugal, Sweden, and the United Kingdom. Finland is excluded from the list of the EU 15 countries.

Offshoring to the rest of Europe Firm has offshored its core business functions to the rest of Europe over the period 2001-2006 = 1, otherwise 0. The rest of Europe includes 12 EU countries (i.e., Czech Republic, Estonia, Cyprus, Latvia, Lithuania, Hungary, Malta, Poland, Slovenia, Slovak Republic, Bulgaria, and Romania) and Switzerland, Norway, Turkey, Russia, Belo Russia, Ukraine, and the Balkan states.

Offshoring to developing countries

Firm has offshored its core business functions to developing countries over the period 2001-2006 = 1, otherwise 0. The developing countries include China, India, South and Central America (including Mexico), and Africa.

Offshoring to other developed countries

Firm has offshored its core business functions to other developed countries over the period 2001-2006 = 1, otherwise 0. The other

24

Table A2 Selected descriptive statistics for the linked data sets

46

TABLE A2

SELECTED DESCRIPTIVE STATISTICS FOR THE LINKED DATA SETS

Firm-level data on outsourcing and restructuring

Individual-level data on outsourcing and well-being

Individual-level data on restructuring and well-being

N Weighted N Mean N Mean N Mean

Outsourcing variables (ISS)

Domestic outsourcing 1096 353 698 0.222 770 0.209

Offshoring 1096 353 698 0.147 770 0.138

Offshoring to the 15 EU countries 1096 353 698 0.222 770 0.209

Offshoring to the rest of Europe 1096 353 698 0.052 770 0.044

Offshoring to developing countries 1096 353 698 0.083 770 0.081

Offshoring to other developed countries 1096 353 698 0.063 770 0.057

Occupational restructuring variables (FLEED)

Task destruction 1096 353 698 0.204 1174 0.189

Task separation 1096 353 698 0.562 1174 0.506

External task separation 1096 353 698 0.394 1174 0.354

Internal task separation (i.e., internal task hiring) 1096 353 698 0.167 1174 0.152

Excess task reallocation 1096 353 698 0.170 1174 0.212

Task churning 1096 353 698 0.714 1174 0.635

The change in the share of interactive tasks 1096 353 698 0.007 1171 0.008

The change in the share of non-routine tasks 1096 353 698 0.014 1171 0.012

Task creation 1096 353 698 0.387 1174 0.472

Task hiring 1096 353 698 0.744 1174 0.789

The change in the share of knowledge workers 1096 353 698 0.022 1174 0.009

The share of knowledge workers in 2006 1096 353 698 0.476 1174 0.461

Well-being variables (QWLS)

Job satisfaction 770 4.048 1174 4.072

47

Uncertainty 770 0.723 1174 0.664

Harm 770 0.281 1174 0.245

Hazard 770 0.397 1174 0.365

No promotion 770 0.418 1174 0.486

No voice 770 0.705 1174 0.677

Discrimination 770 0.377 1174 0.343

Work intensity 770 0.169 1165 0.121

Occupational share variables (FLEED)

Technicians and associate professionals 770 0.216 1174 0.195

Clerical support workers 770 0.110 1174 0.092

Service and sale workers 770 0.078 1174 0.102

Craft and related trade workers 770 0.129 1174 0.164

Plant and machinery operators 770 0.169 1174 0.172

Other workers 770 0.078 1174 0.067

25Outsourcing, Occupational Restructuring, and Employee Well-being: Is There a Silver Lining?

Tabl

e A

3 Co

rrel

atio

n be

twee

n w

orki

ng c

ondi

tion

var

iabl

es

Not

e: n

.s. m

eans

that

the

corr

elat

ion

coef

ficie

nt is

not

sta

tistic

ally

sig

nific

ant a

t the

sta

ndar

d 5%

leve

l.

48

TAB

LE A

3

CO

RR

ELA

TIO

N B

ETW

EEN

WO

RK

ING

CO

ND

ITIO

N V

AR

IAB

LES

Jo

b sa

tisfa

ctio

n U

ncer

tain

ty

Har

m

Haz

ard

No

prom

otio

n N

o vo

ice

Dis

crim

inat

ion

Wor

k in

tens

ity

Job

satis

fact

ion

1

Unc

erta

inty

-0

.195

1

Har

m

-0.1

84

0.21

2 1

Haz

ard

-0.1

62

0.23

5 0.

391

1

No

prom

otio

n -0

.194

n.

s. 0.

139

0.12

4 1

No

voic

e -0

.143

0.

097

0.19

5 0.

199

0.27

7 1

Dis

crim

inat

ion

-0.2

07

0.20

5 0.

153

0.14

7 0.

124

0.12

1 1

Wor

k in

tens

ity

-0.1

55

0.15

5 0.

220

0.22

2 0.

089

0.22

2 0.

158

1

Not

e: n

.s. m

eans

that

the

corr

elat

ion

coef

ficie

nt is

not

stat

istic

ally

sign

ifica

nt a

t the

stan

dard

5%

leve

l.

26

References Abowd, John M., Francis Kramarz, and David N. Margolis. 1999. “High Wage Workers and High Wage Firms.” Econometrica 67(March): 251–333.

Ali-Yrkkö, Jyrki. 2007. “Tuotannon ja T&K-toiminnan ulkoistaminen – motiivit ja onnistuminen.” Discussion Paper No. 1071. Helsinki: The Research Institute of the Finnish Economy.

Ali-Yrkkö, Jyrki, Petri Rouvinen, Timo Seppälä, and Pekka Ylä-Anttila. 2011. “Who Captures Value in Global Supply Chains? Case Nokia N95 Smartphone.” Journal of Industry, Competition and Trade 11(3): 263–278.

Askenazy, Philippe, and Eva Moreno Galbis. 2007. “The Impact of Technological and Organizational Changes on Labor Flows. Evidence on French Establishments.” Labour: Review of Labour Economics and Industrial Relations 21(2): 265–301.

Autor, David H., Frank Levy, and Richard J. Murmane. 2003. “The Skill Content of Recent Technological Change: An Empirical Exploration.” Quarterly Journal of Economics 118(November): 1279–1333.

Bagger, Jesper, Bent Jesper Christensen, and Dale T. Mortensen. 2010. “Wage and Productivity Dispersion: Labour Quality or Rent Sharing?” Royal Holloway, University of London.

Baldwin, Richard. 2006. Globalisation: The Great Unbundling(s). Helsinki: Economic Council of Finland, Prime Minister’s Office.

Bauer, Thomas K., and Stefan Bender. 2004. “Technological Change, Organizational Change, and Job Turnover.” Labour Economics 11(June): 265–291.

Becker, Sascha O., Karolina Ekholm, and Marc-Andreas Muendler. 2009. “Offshoring and the Onshore Composition of Occupations, Tasks and Skills.” Discussion Paper No. 7391. London: Centre for Economic Policy Research.

Blinder, Alan S. 2006. “Offshoring: The Next Industrial Revolution.” Foreign Affairs 85(March/April): 113–128.

Burgess, Simon, Julia Lane, and David Stevens. 2000. “Job Flows, Worker Flows, and Churning.” Journal of Labor Economics 18(July): 473–502.

Böckerman, Petri, and Pekka Ilmakunnas. 2008. “Interaction of Working Conditions, Job Satisfaction, and Sickness Absences: Evidence from a Representative Sample of Employees.” Social Science and Medicine 67(August): 520–528.

Böckerman, Petri, Pekka Ilmakunnas, and Edvard Johansson. 2011. “Job Security and Employee Well-being: Evidence from Matched Survey and Register Data.” Labour Economics 18(August): 547–554.

Clark, Andrew E. 1996. “Job Satisfaction in Britain.” British Journal of Industrial Relations 34(June): 189–217.

Crinò, Rosario. 2009. “Offshoring, Multinationals and Labour Market: A Review of the Empirical Literature.” Journal of Economic Surveys 23(April): 197–249.

Davis, Steven J., and John Haltiwanger. 1999. “Gross Job Flows.” In Handbook of Labor Economics, Volume 3B, edited by Orley Ashenfelter and David Card, pp. 2711–2805. Amsterdam: Elsevier.

Eriksson, Tor. 2010. “Labour Market Outcomes of Internationalization – What Have We Learnt from

27Outsourcing, Occupational Restructuring, and Employee Well-being: Is There a Silver Lining?

Analyses of Microdata on Firms and Their Employees?” In Wage Structures, Employment Adjustment and Globalization, edited by David Marsden and Francois Rycx, pp. 221–243. New York: Palgrave Macmillan.

Geishecker, Ingo. 2008. “The Impact of International Outsourcing on Individual Employment Security: A Micro-level Analysis.” Labour Economics 15(June): 291–314.

Geishecker, Ingo, Maximilian Riedl, and Paul Frijters. 2011. “Offshoring and Job Loss Fears: An Econometric Analysis of Individual Perceptions.” Georg-August-Universität Göttingen.

Griliches, Zvi, and Jerry A. Hausman. 1986. “Errors in Variables in Panel Data.” Journal of Econometrics 31(February): 93–118.

Harrison, Ann, and Margaret McMillan. 2011. “Offshoring Jobs? Multinationals and U.S. Manufacturing Employment.” Review of Economics and Statistics 93(August): 857–875.

Hickman, Daniel C., and William W. Olney. 2011. “Globalization and Investment in Human Capital.” Industrial and Labor Relations Review 64(July): 654–672.

Hopp, Wallace J., Seyed M. R. Iravani, and Fang Liu. 2009. “Managing White-collar Work: An Operations-oriented Survey.” Production and Operations Management 18(1): 1–32.

Ilmakunnas, Pekka, and Mika Maliranta. 2004. “Hiring from Unemployment and Separation to Unemployment.” Applied Economics Letters 11(2): 91–95.

Ilmakunnas, Pekka, and Mika Maliranta. 2005. “Worker Inflow, Outflow, and Churning.” Applied Economics 37(10): 1115–1133.

Ilmakunnas, Pekka, and Mika Maliranta. 2011. “Suomen työpaikka- ja työntekijävirtojen käänteitä: Toimialojen elinkaaret ja finanssikriisit.” Työpoliittinen aikakauskirja 54(2): 6–23.

Kauhanen, Antti, and Sami Napari. 2011. “Career and Wage Dynamics: Evidence from Linked Employer-employee Data.” Research in Labor Economics, Forthcoming.

Lechner, Michael, and Conny Wunsch. (2011). “Sensitivity of Matching-based Program Evaluations to the Availability of Control Variables.” Discussion Paper No. 5553. Bonn: Institute for the Study of Labor.

Lehto, Anna-Maija, and Hanna Sutela. 2009. Three Decades of Working Conditions. Findings of Finnish Quality of Work Life Surveys 1977–2008. Statistics Finland: Helsinki.

Linden, Greg, Kenneth L. Kraemer, and Jason Dedrick. 2007. “Who Captures Value in a Global Innovation System? The case of Apple’s iPod.” University of California, Irvine.

Lurweg, Maren. 2010. “Perceived Job Insecurity, Unemployment Risk and International Trade – A Micro-level Analysis of Employees in German Service Industries.” DIW Berlin: SOEP Papers on Multidisciplinary Panel Data Research No. 300.

Maertz Jr, Carl P., Jack W. Wiley, Cynthia LeRouge, and Michael A. Campion. 2009. “Downsizing Effects on Survivors: Layoffs, Offshoring, and Outsourcing.” Industrial Relations 49(April): 275–285.

Maliranta, Mika. 2009. “Työpaikka- ja työntekijävirrat ja tehtävärakenteiden dynamiikka Suomen yrityssektorilla.” Discussion Paper No. 1177. Helsinki: The Research Institute of the Finnish Economy.

Maliranta, Mika. 2011. “Globalization, Occupational Restructuring and Firm Productivity.” Discussion Paper No. XXXX. Helsinki: The Research Institute of the Finnish Economy.

28

Maliranta, Mika, Petri Rouvinen, and Pekka Ylä-Anttila. 2010. “Finland’s Path to Global Productivity Frontiers Through Creative Destruction.” International Productivity Monitor 20(10): 68–84.

Malone, Thomas W., Robert J. Laubacher, and Tammy Johns. 2011. “The Age of Hyperspecialization.” Harvard Business Review July–August, 56–65.

Munch, Jakob R. 2010. “Whose Job Goes Abroad? International Outsourcing and Individual Job Separations.” Scandinavian Journal of Economics 112(2): 339–360.

Nilsson Hakkala, Katariina, Fredrik Heyman, and Fredrik Sjöholm. 2009. “Multinational Firms and Job Tasks.” Working Paper No. 8/2009. Helsinki: Government Institute for Economic Research.

Østhus, Ståle, and Arne Mastekaasa. 2010. “The Impact of Downsizing on Remaining Workers’ Sickness Absence.” Social Science and Medicine 71(October): 1455–1462.

Sacks, Daniel W., Betsey Stevenson, and Justin Wolfers. 2010. “Subjective Well-being, Income, Economic Development and Growth.” Discussion Paper No. 8048. London: Centre for Economic Policy Research.

Scheve, Kenneth, and Matthew J. Slaughter. 2004. “Economic Insecurity and the Globalization of Production.” American Journal of Political Science 48(October): 662–674.

Statistics Denmark, Statistics Finland, Statistics Netherlands, Statistics Norway, and Statistics Sweden. 2008. International Sourcing. Moving Business Functions Abroad. Copenhagen: Statistics Denmark. Available at http://www.dst.dk/publikation.aspx?cid=13110

Stevenson, Betsey, and Justin Wolfers. 2008. “Economic Growth and Subjective Well-being: Reassessing the Easterlin Paradox.” Brookings Papers on Economic Activity 1:1–87.

Wagner, Joachim. 2011. “Productivity and International Firm Activities: What Do We Know?” IZA Policy Paper No. 23. Bonn: Institute for the Study of Labor.