Microeconometric essays on migration, trust and Satisfaction

Tilburg University

Microeconometric essays on migration, trust and Satisfaction

Bellemare, C.

Publication date:

2004

Document Version

Publisher's PDF, also known as Version of record

Link to publication in Tilburg University Research Portal

Citation for published version (APA):

Bellemare, C. (2004). Microeconometric essays on migration, trust and Satisfaction. CentER, Center for

Economic Research.

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal

Take down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

.1.

TILBURG *1 11T * UNIVERSITY

•1 •

UNIVERSITEIT * * VAN TiLBURG *

BIBLIOTHEEK TILBURG

Microeconometric

_{Essays on}

Migration,

Trust, and

Satisfaction

PROEFSCHRIFT

ter verkrijging vandegraad van doctor aan de Univer-siteit van Tilburg,op gezag vande rectormagnificus, prof. dn F.A. van derDuynSchouten, in het openbaar

te verdedigentenoverstaan van een door het college

voor promotiesaangewezen commissie in de aula van

de Universiteit op

vrijdag 28 mei 2004 om 10 :15 uur

door

CHARLES BELLEMARE

A France et Fran ois pour m'avoir donnd le gout de la vie et de la recherche

A lean-Fran ois,

Acknowledgements

Thisthesiscontains most of the work I pursued during my

four

years stayin Tilburg.

Over this_{time period,manyindividuals contributed either directly or indirectly to the} writing ofthis thesis.

First of all, Ihave_{greatly benefited from}

_my

collaboration

with Arthur

vanSoest who supervised my work from the verystart, andhasco-authored thefinalchapter of

this thesis. I thank him for havingencouraged me topursue my ownresearchideas, andfor having given so much of his time, guidance, and words

of

encouragement.

Arthur's joy

toinstructothers, and his disinteresteddedication tohis students, sets an example

which I will try

to emulateallthrough myacademic life.

Duringthefirstsemester ofmystudies, I wasconfronted withthe technical_aspects of econometrics for which I was not reallyprepared. Making this transition

would

have beenmore _{difficult without the help of}Betrand Melenberg. The _{extra evening}

lectures he gave me on thewhite board of my office willhavegreatlyimproved my understanding ofthefieldof econometrics. Bertrandco-authored the last_{chapter of} this thesis, and I amdelighted thatheaccepted to join

my

thesiscommittee.

Special thanks gotheother members ofmy thesis committee, Claude

Montmar-quette,Christian_{Dustmann, and Jan van Ours.}

My

discussions with them have shaped the way I

think

about economic problems. I am especiallygratefulto Claude

Montmar-quette

for

havingintroduced me toexperimentalandbehavioraleconomics while I was working attheCIRANObefore entering the Ph.D.program

in

Tilburg. Hisenthusiasm

forthesefields has beenasourceof_{greatinspiration. I look forward}tocollaborate with

him and his team at theCIRANO in_{the years to come.}

I am grateful to CentER and to theDepartment ofEconometricsfor havinggiven me

the possibility topresent my workat numerous conferences,workshopsandsummer schools. I have gained muchfromtheseexperiences,both personallyand

profession-ally. I would also liketothank Marcel Das and all the teamatCentERdatafor having

Viii

My

stayin'Illburg hasbeen_{highlighted by my}

_joint

_venture

_with

_{Steffan Berridge} in runningtheinfamousMieredikhof Hotel. _{Memorable cocktail parties and backyard}

fires have helped us andfellow students to remain cheerful_{through the cold winter}

nights

of

111burg. I willespecially remember Steffan's

mild

chilies, his barbecue

black-out, and his special way

of

havingwhiskey.

_{My visit}

to_{China with him and Tu Qin}

has left me withawealth

of

memories. It has beenapleasure to know him and I am honored thatheaccepted to be one of_myparanymphs.

I would also like tothank my twooffice _{mates, Joost}Driessen and in particular

Pierre-Carl _{Michaud, for stimulating conversations anddebates.}

_My

_{thanks extend}

to all thePh.D. students and friends in

lilburg,

with

specialmention _{to Pierre-Carl,} Antonis, David, Tu_{Qin, Youwei, Dantao, Steffan and Vera, who have allhelped me}

balance mylife between workand leisure.

It does not take much reflection to_{realize how much our}achievements in life

de-pend on the_{support given to us by our family. Leavingmy}

parents and brother has unquestionably been the most

difficult part of

mystudies abroad. I rejoicemyself in the thoughtofcoming back home, and it is

with

great emotion that I dedicate_mythesis to them.

Finally,Sabine hasshared mylife throughoutthe

writing of

this thesis. Her care,

comfort,and_presence,

will

remain the_{most importantheritage of my timein Tilburg.}

Charles Bellemare

Contents

Acknowledgements Vii

Contents ix

1 Overview ofthethesis 1

2 IdentificationandEstimationofEconomicmodels_{of Outmigration using}Panel Attrition 9

2.1 Introduction . . . , . . . , , , , . . , . . . 9

2.2 Identificationof outrnigrationparameters . . . 12

2.3 Parametricmodelandestimationmethod . . . 16

2.4

Data... 20

2.5 Results and simulations . . . . . . . . . . . . . . 22

2.6 Conclusions . . . ... 29

Appendixto Chapter 2 30

3 A Life-Cycle Model of Outmigration and EconomicAssimilation of Immi-grantsin Germany 41

3.1 Introduction . . . .4 1

3.2 Economic

m o d e l. . . 45

3.3 _{Estimation procedure . .. .. . . ... . .. . . .. 49}

3.4 Background and Data . . . 56

3.5 Estimationresults for thestructuralmodel . . . .. . . . .. 58

3.6 Conclusions . . . 66

Appendixto Chapter 3 68

X

4 OnRepresentative_{Trust 81}

4.1 Introduction . . . .8 1 4.2 TheExperimental Design and the

Sample ... 85

4.3 Resultson_{Trust . . . .} . .

. . 88

4.4 Results on

Trustworthiness . . . 95

4.5 Results on_{Participation} .

. . . .9 9

4.6 Discussion and

Conclusions . . . , . . . 100

Appendix

toChapter4 ₁₀₃ 5 Semi-Parametric_{Models for Satisfaction withIncome 115} 5.1 Introduction _{. . . , .} 115 5.2 _{Models, EstimationTechniques,and SpecificationTests . . . 117}

5.3 Data and

Variables . . . 125

5.4 Results . . . .

_{. . . 127}

5.5 Conclusions...···

₁₃₁

Appendix

toChapter5 ₁₃₂

Bibliography ₁₄₃

List

of

Tables

2.1 Panelattrition forWest-German and_{Immigrantsamples}1985-1999. . . 31

2.2 _{Descriptive statistics} for West-German and _{Immigrant samples, 1985}

and 1999 . . . 32

2.3 Unemployment Rate per Wave and Land1985-1999 . . . . . . . . . . 33

2.4 Covariancestructure of thetimevariant and timeinvariantcomponents 34

2.5 Estimation resultsfor o u t m i g r a t i o n. . . 35

2.6 Estimation resultsforincome and_{work equations . . . . . . 36}

2.7 Simulation

Results . . . 37

3.1 Descriptive

statistics, 1985 and 1999 . . . . . . . 69

3.2 Estimation results of thestructural

model ... 70

3.3 Panelattrition

for

West-German and_{Immigrantsamples}1985-1999. . . . 71

3.4 Simulatedmigration

durations . . . 72

3.5 Estimation results of the reducedform model ... 73

4.1 _Descriptive

statistics . . . . . . . . . . . . . . 108

4.2 Sender

results . . . . . . . . . . 109

4.3 _{Responder results}

-

Tobitestimator . . . 110

4.4 _{Responder results}-STLS-estimator . . . 111

5.1 DescriptiveStatistics of thesample . . . 133

5.2 LM SpecificationTestsOrderedProbit . . . 134

5.3 Estimation Results, Extended_{Specification . . . . . . . . . . . . . . . . . 135}

5.4 _EquivalenceScales . . . 136

List of

_Figures

2.1 _{Logmonthlyearnings and}work_propensities-_{Germans andimmigrants 38}

2.2 Simulation

Results . . . 39

3.1 Proportions

of

immigrants

working

inGermany,not

working

and attri-tion per time period,

1987-1999. . . . . . . . . 74

3.2 Goodness of fit of themodel ... 75

3.3 Simulateddistributions fortheforward looking

model . . . 76

3.4 Simulateddistributions forthe_{myopic looking}

m o d e l. . . 77

4.1 Distribution

of

experimentaltrust. . . . 112

4.2 Returnratio

of

_{responders for each}units_{received, strategy}method. . . 112

4.3 Estimated_density

of

_potentialreturns oninvestments . . . 113

5.1 Simulated rejection probabilities Fan and Li (1996) test . . . 137

5.2 Distribution

of

satisfaction

with

income . . . 137

5.3 Distribution of loghousehold income_byhousehold size . . . 138

5.4 Nonparametricregressionofsatisfaction

with

income onlogincome by

household size . . . 138

5.5 _{Non-parametric}estimate oflink function, with95%uniformconfidence bands.Ichimura'sSLS;_{specification 2 . . . 139}

5.6 Non-parametricpartpartiallylinearmodels . . . 139

5.7 Non-parametricpartgeneralizedpartially linearmodels . . . 140

Chapter 1

Overview of

the thesis

This thesis presentsfouressayswhich illustratetheprocessof makingmicroeconomic inferenceson_{economicallyrelevant parameters characterizing}

individual

welfare and decision_{making using}

individual

andhousehold data.This_{process isusually thought} of asathree_{stepapproach,} whichstarts_bydrawingoneconomic_theoryto focus on the parametersofinterest.Giveneconomically relevant parameters havebeendefined, the second step of theprocessconsists of obtainingan "ideal" datasetwhich is rich

enough to

identify

and estimatetheseparameters. Itisoften the caseineconomics and

otherobservational sciences that this"ideal" data set isnot available,whichcanoften jeopardizeidentification of some or all parameters. In thiscase, researchers_typically trytocircumventthe_problembyeither_{imposingassumptions on}individualbehavior,

or bygettingbetter data. Ofcourse, in theformercase,the_{credibility of}theinferences depend on the realism andfalsifiability oftheassumptions made,whilethefeasibility ofthelatter_{approach depends on}whether or not onecan_actuallycollect richer data. Once identification hasbeenachieved, thefinal_step consists_{of using existing or new} estimatorsin orderto extract therelevantinformation from thedata. Becauseseveral estimators,under different set

of

assumptions,cansuccessfully extractthis informa-tion, empirical researchers havea _degree offreedom to choose the model which is

estimated. Fortunately, economictheoryoften restricts the setofpossible econometric models, andmodeltesting procedures and goodness of

fit

evaluationscansometimes be used to test the

_validity

of maintained_{assumptions,providing a way}toincrease the

reliability oftheinferences made.

The essays in thisthesis arepresented asindependent chapters, and analyze the characteristics anddecisionrules_{of immigrants who}leavetheirhost country, evaluate

2 Chapter 1. Overview of the thesis

the determinants of trust andtrustworthiness for different sub-groups ofthe

popula-tion, and estimate household equivalence scales. Ineachchapter,identification of the

economic parametersofinterestis_{achieved either byimposingrealistic and falsifiable}

assumptionson

individual

_{behavior, or}

by

_{collecting better data. Taken together,these}

essaysdevelop a widerangeofeconometric modelsin order_{to estimate the}

parame-tersofinterest,fromdiscretechoicestaticand panel data models, tostructural dynamic programming_{models, nonparametric models, and semiparametricmodels. Although}

issuesof identificationandestimationarediscussed inallessays,theirrelativeweight

differsacross_chapters.

The firsttwo chapters deal

with

outmigration,defined as thedeparture of

immi-grantsfrom their adoptivehomeland._{During the}lastdecade,economists have increas-inglybecome aware _{that outmigration is not only}a world wide_{phenomena, but the}

magnitude oftheoutflows insomecasesbe sizeable. Within_{Europe, the case of}

Ger-many stands outprominently, withanestimated halfa

million

_{immigrantsleaving the}

countryevery year (OECD,2001)._{Understanding}themotivations

for

these_departures

is an_{important input for policy makers whomust forecastoutflows}

_of

immigrants in orderto adjusttheir immigration policies to fitthefutureneedsof theirlabor markets, and to assess the economic performance _{of immigrants who remain in the country.} Several theories havebeenproposed toexplainthesemovements, someof which pre-dict thatthe_relativelysuccessful_immigrantsleave_{their adoptivecountry,whileother}

theoriespredict that most ofthe departures areby unsuccessfulimmigrants. Some-what surprisingly,there hasbeen

little

structuralinference on the preferences, _tastes, and _{other primitivescharacterizingimmigrants which would allowtovalidatethese} theories, helpusunderstand the typeof immigrants whichself-select themselves out of the country,and

would

enableinteresting policyexperiments. Given thatthe

param-eters ofinterestare clearly_{defined, the lack of}inferences can betraced back to data

limitations. The body ofdatatypicallyavailable toconducttheseinferences usually

consists ofasample_{of immigrants, reinterviewed in subsequentyears. This type of}

surveydata rarely contains_{indicators of immigrantsdepartures. Instead, what is} em-piricallyobserved isattrition from_{the panel,}caused either_{by immigrantdepartures} or by non-response of the nondeparting immigrants. Thus, thedata fails todirectly

identify

the parametersofinterest, and gatheringthe missing information

would in

many instancesbe_{prohibitivelycostly.}

im-3

posed on

individual

decisionmaking in order tononparametricallyidentifyseveral

important economic parameters characterizing the performance

of

outmigrants, no-tably theconditional outmigration probability, theconditional

work

probability, and

theconditional expected labormarketearnings

of

outmigrants. Theassumption

re-quired isthe existence_{of immigrants} whose observable characteristics are such that

their outmigration probability isclose to zero. Equally important,becausepanel

at-trition is, contraryto outmigration,always observable in surveydata, these results open up thepossibility tomake inferenceswhich werenotpreviouslypossible due to

data limitations. Toextract theparametersofinterestfrompanel attrition, the

chap-ter presentsa

limited

_{dependent variable model} for_{panel data}

with

randomeffects,

wherelabor market_earningsareestimated_{jointly with both work}and _outmigration equations. An importantaspect of theempirical model is that itallows to test the key

identification assumption ina_{simpleway. Indeed, the results indicate that}a

substan-tial number of immigrants haveanoutmigration probabilityclose to zero. The main

results fromtheempiricalanalysis show that outmigrants arenegativelyselected in terms of labor market_{earnings and}

work

_{propensities. Throughout}the _sample pe-riod, we find that the gap ineconomic performanceof immigrants who remained in Germanyand thosewho outmigrated was sizeable,

with

outmigrants having work probabilitiesand expected labormarket_{earnings respectively 18% and}30%lower than immigrantswho remain inGermany. Finally, the estimated outmigration rate in the

samplepopulationisfound toberoughly 3% per year.

Chapter3_{presents and estimates}a _{structural dynamic model} where_immigrants

simultaneouslychoosetheir workandmigration durationsover theirlife-cycle.

Con-trary to existinglife-cycle models,immigrantsareallowed toface_uncertainty about

their future labor market outcomesall through their migration experience. Because outmigrationis _rarely observed at the individual level, structural estimates of

life-cycle models

of

outmigration have yet tobe_formallytested._{This chapter draws on the} identification results oftheprevious_{chapter and extends the approach to}adynamic programmingsetting. This allows to make direct structural inferenceson_outmigration

life-cyclebehavior

without

havingto observeoutmigration and to testthe

validity of

several existinglife-cycle theories. Our findings confirmthe hypothesis recently put forward intheliteraturethatoutmigration isnotentirely driven byearnings

differen-tials. _{Specifically, we find}that_{immigrants who}feel_{integrated in}the German_society,

na-4 Chapter 1. Overview of the thesis tive country are less

likely

tooutmigrate. The results ofthis chapter also

highlight

the importanceof incorporating the

work

decisionalong withthemigration duration

decisionof immigrants,afeaturepreviously ignored intheoutmigration literature. It is found that both immigrants with relatively low and highlabormarket_experience

havea_{greater overallutility}to_{outmigrate, whichsuggests a}U_shaperelationbetween

labor marketexperience and theoverall

utility

to outmigrate. The decreasein overall outmigration

utility

starting from lowlevelsofexperienceisconsistent

with

increas-ingpsychologicalcosts associatedwith outmigration,keeping earnings constant. The convex increase

in

overall outmigration

utility

predicted to occurbeyond25years of

labor market_experienceisconsistent

with

_{progressively}lower_{psychological costs of} outmigrationanddiminishingreturns to labormarketexperience in the host country.

Theseresults are interestinggiven that most of the outmigration literature has

ana-lyzed outmigration

within

an_{earningsdifferential paradigm which, byconstruction,} orient policy recommendations towards measures aimed atinfluencing theearnings

differentialbetween the host and thenew destination. The_{results of this chapter do not} rule outtheimportantrole playedby labormarket earningsin determining migration durations. However, they doindicate thatthe shape of themigration duration distri-bution isalsodetermined by pastworkdecisions,indicating that much canbegained

froman_analysisin whichlabor_supply decisionsare_endogenously determined. The

chapteradditionallypresentsaseries_{of policy}simulations aimed at_{evaluating how}

themigration duration distribution of immigrants is affectedby government policy

aimedat_{speedingup economic integration of newly arrived immigrants. Itisfound}

thatmeasuresaimed atimprovingaccess to thehostlabor

market

improving

acquisi-tion

of

_{language skills, and}lump-sum taxation willhave

little

impact ofthemigration duration distribution ofa_{representative immigrant, but}havesubstantialinfluence of migration durationdecisions of

low

income_immigrants.

In chapter 4, the economic parameters

of

interest arethedeterminantsof

individual

trust andtrustworthinesspropensities. Compared to theprevioustwo chapters,

iden-tification oftheseparametersisachieved, notby imposinga

priori

assumptions on the

individual

_{decision maker, butbyusingbetter data in the form of}a random_sample oftheDutch population who playanexperimentalgamemeasuring

individual

trust andtrustworthiness_{propensities, both}of which havebeenshown tobe_strongly corre-lated

with

economic_growth.Theidentification_approachis_{novel in many respects and}

5

experiments which uses a sample

of

subjectswhich are representative ofa nation's

population. Substantial heterogeneity

in

trust and trustworthinessbehaviorisfound intheDutch population. Of particularinterest are the rolesplayed bythe education level and the ageofsubjects. The chaptershows that both ofthesebackground

char-acteristics playan important role in determiningtrust andtrustworthiness, although

they affect trust and trustworthiness in very different ways. While the inverted U

shaperelation between trust and ageusually

found in

the literature is also present

in our data,

with

trust increasinguntil the age of 30and decreasingbeyond that, the relationbetween trustworthiness and ageisfound to beU shaped,

with

trustworthi-nessdecreasing until the age of 30, andrisingagain beyondthat point. Theeducation patterns are also very different. We find an inverted-Ushaperelationbetween edu-cation and trust, and a U shaperelationshipbetween education andtrustworthiness. Such_{opposite influences}ofbackground characteristicsarepuzzlinginsofar as it is

typ-icallyassumed that trust andtrustworthiness go hand inhand,which

would

suggest

that botharedeterminedin similar ways. Asa secondmain contribution, the chapter presents a new andsimple wayto compare theinformationalcontent ofbothrevealed (experimental) and stated (survey) trust measures. The literature has until now

as-sessedthe

validity

ofsurvey trust questionsby testingwhether or nottheypredicted

well

_experimentaltrust. One of themain_{messages ofthe chapter is that this}method of validation has_{been given too much}attention, primarilybecausethe_{predictive power}

of surveymeasuresisintimately linked tothe sample used, the amount ofbackground

informationavailable on the subjects, and theexperimental design. Itisshown that by carefullyselecting samples and designs, experimentersincreasetheir odds of finding

either a low or high predictive power ofthesurveytrustmeasure. Thus,despite that

contrary totheexisting literaturethesurvey trustmeasure used in this chapter predicts well trust intheexperiment, this is not takenasevidencevalidating the use of survey

trust questions. Withregards to theinformationalcontentbothmeasurescarry on the

determinants oftrust,thedifferencesare_{compelling. It}isshownthateducation has an

inverted-Ushaperelation

with

experimental trust while it does notcorrelate at all with

answers to the statedtrust_question. Incontrast,religioncorrelates wellwithanswers tothestated trust question, but not at all

with

experimentaltrust.

In_{stark contrast, chapter}5 _presents a_{example where statedresponsesprovide a}

simple solutiontoidentificationproblemswhichcannotbeovercome using data on

6 Chapter 1. Overview of the thesis

incomewhichshouldbe_{given to}ahousehold witha_{specific composition}ofadults and

childrentoreach the samelevelofwelfare thanareference household.These_quantities are important for public policyconcerningwelfarebenefits andchild_{allowances, and}

have also been used to analyze incomeinequality

within

and between countries and

for_{the analysis}

of

_poverty.Thetraditional_{approach to estimationofequivalencescales} usesconsumerdemand_{systems,relying}onvariation in_{expenditure on commodities}

acrossfamilieswith different_composition. Pollakand_{Wales (1979)}andBlundell and

Lewbel(1991) demonstrated that consumer demand data alone is not_{richenough to} nonparametrically identifythe_{level of the equivalence scales. In this chapter, the}

fail-ure to identify levels using revealed consumption choices is overcome using stated

responses to questions on satisfaction with _{income, which directly reveal} a cardinal

measureofconsumer_utility. Stated_{answers to such questions provide empirical}

re-searchers with a _{discrete and ordered response variable,usually}

analyzed using

or-deredprobitand linearregression _{models. The emphasis of the chapter is to recover}

estimates ofthelevelof_equivalencescalesforGerman households_usinga_{broad range}

of_{estimation procedures, from the usualorderedprobitandlinearregression models,}

to more recent nonparametric andsemiparametricmodelswhich_{weaken parametric}

assumptionsof linearand ordered _{probit regression models. Given the broad range}

of_{possible estimators,} the chapter uses consistent waysof testing each underlying model _{assumption against general}forms

of

_{mis-specification. The}main _{findings of} the chapter arethatorderedprobitmodelsarerejected,whileseveralsemiparametric estimators are not. Despite this, estimates of the equivalencescales arefound to be

ro-bustacrossmodels. _{Finally, it}isfound that_equivalencescales_{monotonically}increase

with

thenumber of children for low and

middle

_{income households, but}arefound to be relatively flat for highincome households.

Collectively, these essayspresent waystoovercomeidentification problems in

mi-gration, experimental economics, and measurementof_{household equivalence}scales. Ineach chapter,illustration ofthe approachesrequired that theybeapplied either to veryspecific countries,institutional_{settings, and choice environments.} However, the usefulnessofthese essays goes beyond theseapplications, andprovidemeans for a

significant number of extensions. Theidentification _{strategy proposed}in _{chapter 1}

opens up thepossibility todoresearch onthe economicsof outmigration in practically any country with an_{ongoing panel surveying immigrants. Because}

outmigration is

aworldwide_{phenomena, this avenueoffutureresearch is}

or-7

der to improve our understanding ofthe flows of human capital takingplace across

the globe. Moreover, because this strategy is relatively general, itcaneasily be ap-plied to analyze any form of migration movements, fromthe departures ofnatives, to movementsofindividualswithina_country Theresults presented in thefirst two

chapters of this thesis show that such extensions havea_{promising future.} The thesis also showed that one cannow_{successfully makepopulation}inferences onbehavior in

experimentalgames. Whilethe thesishasshown that inferences based onsurvey

re-sponses can insomecasesdiffer remarkably fromthose basedon revealed preference data collected_{from experiments, it has also}been shown that_surveydata remains a useful source toidentifyeconomic parameters related to measuresofhousehold sat-isfaction. _{More generally, it} is_{hoped that} this thesis has alsodemonstrated that the combinationof_{survey andexperimental}methods has the_potentialto_provideuseful

insights in many differentgames,settings, andpopulations, which

will

undoubtedly

Chapter 2

Identification

and

Estimation of

Economic

models

_{of Outmigration}

using

Panel

Attrition

2.1 Introduction

The widespread and often sizeableflows

of

immigrants leaving theiradoptive home-lands,which we

will

refer toas_{outmigration,have recently received a lot}ofattention. The case of Germany is a particularly revealingexample, withan estimated yearly outflow of halfa

million

_{immigrants over thelast decade (OECD,2001).} Several the-ories have beenput forward to motivate outmigrationmovements. Theories based

on earningsdifferentialsbetween the current and new_{destination (Harris}and Todaro,

1970),higher marginal

utility

of

consumption in thehome country (Djajic and

Mil-bourne, 1988),highreturns tohumancapitalinvestments in thehost country

(Dust-mann, 1993), information dissemination(Stark, 1995),credit market rationing in the native country (Mesnard, 2001), and several sociological factors suchas

_family

uni-fication, health satisfaction, feelingsof_{being integrated} in_{society (Stark, 1998) , and} the qualityandproductivity ofanimmigrant'ssocialnetwork(e.g. Carrington, Detra-giache and Vishwanath, 1996).

Thesetheories do not

trivially

predict a specificcomposition of departing

immi-grants. Itcould be the caseforexamplethateconomically successfulimmigrants with

a_{relatively higher marginal}

_utility

_{of consumption in their native country might opt}

to leave despiterelatively lowerearnings inthatcountry,while persistently

10 _{Chapter2. Identification ofOutmigration}

ful

immigrantsmight find aworthwhile to keepon searchingforbetter labor market prospects and move to anew destination. The lack ofaclear-cut theoreticalprediction

concerning theeconomic performance

of

outmigrants complicates the design of im-migrationpolicieswhichareoftentailoredaround thenotion thatlabor market

equi-libriums may be affected_{by outmigration flows of non-randomlyselectedworkers.} Careful forecasting ofthequality ofthemigration flows isthusnecessary

if

immigra-tion policies are to meetthe futureneeds of the labor market. Additionally, recent

theoretical andempiricalevidencehassuggestedthatmeasuresofeconomic progress

of immigrants in theirhost country canbe_{adversely affectedbyselectiveoutmigration}

(Schultz, 1998, andEdin, LaLonde, andAslund,2000).Policies aimed atimproving the

labormarket integration ofitsimmigrant population may thus alsobemisguided if it

reliesonthesepotentiallybiased measuresof immigrant assimilation.

In this_{chapter, we}areinterestedin_{testingcompetingexplanationsof outmigration}

decisions while at the same time_{assessing the}robustnessofmeasuresofeconomic

as-similationto_outmigrationflows. We do so_byestimatinga_{limited-dependent}variable panel data model where labor market earnings,workdecisions, and_outmigration de-cisionsare_{jointlydetermined and depend on earningsdifferentials, family unification,} creditmarket_{rationing,health satisfaction, and feelingsof beingintegrated in the host}

society. Unobservable characteristics such as an_{immigrant's ability or}the_{quality and}

productivity ofhis socialnetworkare_{incorporated in}themodelasindividual_specific

unobserved heterogeneity components. The generalerror structure ofthe model al-lows totestwhether outmigrants areself-selected in terms

of

labor marketearnings andin terms of

work

status. Themodel alsoallowstoassess theimpactof outmigra-tion selecoutmigra-tiononestimates ofmeasuresofeconomic assimilationrates. _{Compared to} existing empiricalmodelsof outmigration(seebelow),ourmodel hastheadvantage

of both incorporating thedecision to work in a _{natural wayand characterizing the} relationshipbetweenworkstatus and_{outmigration.}

Estimation of our model_{requires panel data onimmigrants followed over a} rela-tively longperiod of time. As Dustmann(2002)recentlypointed out,interesting

iden-Section2.1.

_{Introduction 11}

tify

the economic parameters characterizing the performanceof outmigrants

without

observingoutmigrationdecisions. Jassoand Rosenzweig(1990)

identify

thedirection of outmigrationearningsselectivityby comparingthe_{skill composition ofspecific}

co-horts over time. Hu (2000) andLubotsky(2000)estimate the parameters of the earnings function_{of immigrants who remain in}the_{country,controlling for non-random} outmi-gration selectivity by matchingcross-section data setsand

longitudinal

socialsecurity

earnings records. Theseapproaches are not

without

their own limitations-they do not

identify

thelevelofearningsofoutmigrants,censusand earnings records often have

little information on boththehuman _{capital level and sociological characteristics of} migrantswhicharerequired to testoutmigrationtheories 2, they do not easilyallow migration duration _{decisions to depend on unobserved characteristics suchas} inher-ent ability orthe_{quality of}an_immigrant'ssocial_{network which,}at leaston theoretical grounds,are_importantdeterminants_{of migration durations.}

Inthis chapter, we presentconditions under whichtheoutmigration probability,

the

work

probability, andthe expected labor market earnings

of

outmigrants are,

con-ditionallyon observable characteristics,nonparametrically identified fromsurvey data

with

_sampleattrition.The cornerstone ofour identification_{approach consistsof using}

panelattrition asaproxyvariable for outmigration,and subsequently separating

at-trition which is not duetooutmigrationmovements from realoutmigrationdecisions.

Our_{approach overcomes several of the shortcomings}

of

earlier approaches used to

identifythe economic performanceof outmigrants.First,our_{approach is to}our knowl-edgethefirstonewhich hasthe_potentialto_{provide nonparametric identification of the} economic performance and movementsof outmigrants.Second,because_{the approach}

proposed uses _{survey panel data instead of}census _{data, unobserved heterogeneity}

can easily beintroduced in themodel. Third, _{the approach}is _{general enough to be} applied toany

country with

an ongoing panelof immigrants (examplesofcountries

with_{such panels}are_{Canada, Mexico, Germany, and the United-States.) and can be}

easily extended to estimatemanydifferenttypesofeconomic modelsof outmigration.

An example of_{this flexibility is our ability}to analyze theinteraction_{between earnings,} work,and _outmigration decisions ina unifiedframework which wasnot _previously possible

without

observing actualmigration_{decisions. Finally, the estimator proposed}

is simple to apply, andissimilar in

spirit

to estimators proposed to deal

with

12 Chapter 2. Identification of Outmigration

We estimate our model using thepublic use file of _{the GSOEP. We use data on}

native Germans as a reference group to compute earnings_{assimilation rates for our}

immigrantsample. We show how theavailability ofasampledrawn from thenative population has theadditional_{advantageof providinganatural way to test some of the} identifyingassumptions ofthemodel.We_{estimate the annualoutmigration rate in our}

sample tobe_{roughly 3% per year over our time horizon. Our simulation results}

indi-catethat_{averagelog earnings}

of

_outmigrantsremained_roughly18%lowerthanthose

of immigrantstayers,aclear_{indication that outmigrantsaredrawn fromthebottom of} the incomedistribution. Moreover,outmigrants areshown to have

work

probability

25% to45%lower than thatofimmigrantstayers over theperiod_{considered. Finally,} we do not find thatassimilationrates areparticularlysensitivetooutmigration, which

contrastswithexisting resultsfound inthe literature.

The rest of this chapter isorganizedasfollows. Section2.2_presentsourapproach

to identifytheeconomic parametersofinterest. Section2.3presents the econometric model used to_{model outmigration in conjunction with the}

_work

_{decision and labor} market_earnings. Section 2.4_{presents the data used in the chapter. Section2.5 discusses}

theempirical _{results of the model and tests for thepresenceof outmigration bias. It} further_{presents some}simulation results usedtoevaluate the fit ofthemodel and to quantifythe economic performanceof outmigrants.Section2.6concludes.

2.2 Identification of outmigration

parameters

Each immigrant ofapopulation living in thehostcountryischaracterized in agiven

time period by thevector (w, p, ru, x, z,_{s) where}w denoteshispotential labor mar-ketearningsconditionalon characteristics x, p isabinary variable takinga value of 1

whenthe_{immigrant works}andwhose outcomeisconditioned onavectorof observ-able characteristics z, ru isabinary indicator takingavalue of 1 when theimmigrant

outmigrates in the next time period andwe conditionthis outcome onavector of

char-acteristics s. Wedefine X as the _{vector of all distinctelements of (x, z, s). We are}

interested in makinginferences on Pr (p = 1 lru = 1,

_X),

_conditional

_work

_probability

of an outmigrant, and on E {wlp = 1, ru = 1, X},_{theconditionalexpectedearnings of} outmigrants. Theinferential_{problem consistof identifyingthesequantities when,}

in-steadof_{observingoutmigration,we observea}

proxyvariablero,_panelattrition, which

Out-Section2.2.Identifcation results 13 migrationand

attrition

arerelatedbecause an_{immigrant who}leavesthe_{country must}

also leavethe panel

with

probability 1.3

We illustrate theidentificationproblems for thecasewhere we want tomake

infer-enceson E{wlp = 1, ru =

_1.X}.

However,allresults extenddirectlytoidentification of

Pr (p = liru = 1, X).4 Theconditionalexpected labor marketearningsof immigrants

who leave the_{panel in the next timeperiod can be expressed,usingiterated}

expecta-tions, as

E{wlp = l.ro = 1,X} = E{wlp =l, r° =l, ru =l,X} ·Pr(ru =llro = 1,X)

+E{wlp = 1,78 = 1,ru = O. X} · Pr(ru = 0170 = 1. X) = E {wlp = 1, ru = 1, X} ·Pr (ru = 1 lro = 1, X)

+E {wlp = 1, ru = 0, X} · Pr (ru = Olro = 1,

_X)

(2.1) wherethesecond_equalityfollows from the fact that once we know ifan_{immigrant left}

or not the country, observing whether he left of not the panel doesnot_{contain any} addi-tional information onthe earnings ofthisimmigrantconditional on X.5 Animmediate

consequenceofequation (2.1) is thatusing panelattrition asa_proxyvariablefor

outmi-gration

in itself will

give biased and inconsistent estimates of theconditionalearnings

of outmigrants. This is sobecause the conditional_{expected earnings of immigrants}

who leave the_{panel will ingeneral be}a_{weightedaverage of the}conditional_expected

earnings of outmigrants mixed withthe conditional earningsof immigrants who

re-main in the host country.Themixing probabilitiescontrol the size of the bias. The key

parameter is Pr (ru = Olr° = 1,

_X),

whichrepresents theprobability thatanimmigrant

stays in the thehostcountrygiven that heisobserved to leave the panel. Thehigher this probability, the higher will be the bias.

If

everyimmigrant wholeaves the panel

also leaves thecountry, r°would perfectlymeasureoutmigration, Pr (ru = Olr° = 1, X)

wouldbeequal to zero, and the bias would be zero.

Next, wefollow thesame stepstoderivetheconditionalexpected earnings of im-migrantswho remain inthe panel

E{wlp = 1, ro = O,X} = E{wlp = 1, ru = 1, X}·Pr(ru =llro = O,

_X)

(2.2)

+E{wlp = 1, ru = 0,X} ·Pr (ru = Olro = O, X)

Becausean_immigrantcannotbeobserved to have left the_{countrygiven he}isobserved

to be in the panel, Pr (ru = 1 Ir° = 0, X) = 0, Pr (ru = 0 Ir° = 0, X) = l and _(2.2)

simpli-fies to

14 Chapter 2. Identification of Outmigration whichindicates thattheconditionalearningsof immigrants who remain inthe panel

coincides withthe expected earningsof immigrants who remain in thehost country, andit follows that E {wlp = 1, ru = 0, X} isnonparametrically identified from data on immigrants inthe sample.Substituting (2.3) in (2.1)weobtain

E{wlp = 1,f = 1,X} = E{wip = 1, ru - 1, X}·Pr(ru =llro = 1, X)

+E{wlp = 1, ro = 0, X} ·Pr (ru = Olro = 1. X)

which canbesolved interms of E {wlp = 1, ru = 1, X},theparameter we hope to

iden-tify,

E{wlp=l, ru=l,X} = E{wlp=l, ro -1, X} ·Wi(X)-1

-E{wlp = 1,t - 0, X}· Wo (X) Wi (X)-1 (2.4)

Equation(2.4)shows that theconditionalexpected earningsofoutmigrants can be ex-pressed as a _{weighted average of}two conditional _expectationswhichareidentified from the data. Iftheweights canbeidentified, thentheconditionalearnings

of

outmi-grants willbe_{identified. ApplyingBaye's rule, theweights}are_{given by}

Wo (X) = Pr (ru= Olt =1, X)

= Pr (ro = 1lru = 0, X) Pr (ru = OIX)

Pr (r° = 1 IX) Wl (X) = Pr (ru = 1 lro = 1, X) - Pr (P = 1 1 r u - 1 X)Pr (ru = 1 IX) v ' , Pr ( r°= 1 IX) =1 Pr (ru = 1 IX) Pr (ro= 1IX)

Pr (r° = 1 IX) is identified fromthe attrition data. What remains to beidentified is

Pr (ru IX) and Pr (ro = 11ru = 0, X). Itisclear assumptions mustbeplaced on the data generatingprocesstoidentifytherelationshipbetweentheobservable

attrition

indica-tor r° and the unobservableoutmigrationindicator ru.To

simplify

thenotation, we will

denote a10 (X) E Pr (r° = 1 lru = O.

X)

Using the law oftotalprobability,theattrition

probability canbe expressedin general terms as

Pr (ro = 1IX) = alo (X) + 11 - alo (X)] · Pr (ru =

lIX)

(2.5)

and theprobabilityofremaining inthe sample as

Pr (r° =

_{OIX) =1- {alo (X) t[l- trio}

_(X)]

Pr (ru = _1IX)}

Section2.2._{Identification results 15}

Several assumptions canbe_{imposed on the data generatingprocess}in orderto

_identify

both alo (X)andPr (ru = 1IX).

Assumption Al a10 (X) = alo 2 0

Proposition2.1. _{If Al holds}andthereexists a X such that Pr (ru = 1 IX) = 0. E {wip = 1, ru = 1, X} is nonparametrically ident(Red.

Proof. Given the foregoing discussion, it suffices to establish that theconditions of

theProposition

identify

the weights Wo (X) and Wi (X). From (2.5), it follows that

Pr (r° = 1IX) = 0:10whichidentifies 0:10from

limit

observations satisfying Pr (ru = 1 IX) =

0. Given a lo isidentified, Pr (ru = 1IX)isidentified from(2.6),which implies that the

weights Wo (X) and Wo (X) are bothnonparametrically identified. 0

Proposition 1 shows that

if

attrition which is not duetooutmigrationisrandom in

the population (Al),

all importanteconomic parameters characterizingoutmigration

behavior canberecovered from the data

if

there existsa_sampleof_{permanentmigrants;} i.e. immigrantswhoseoutmigrationprobabilityisclose to zero. Inpractice, this does not seem to bea_{strong requirement, especially}

for

western countries where_permanent

migrationisknownto occur at avery_{large scale (OECD, 2001). Note that}inpractice,

Al needs not to hold

if

attrition which isnotrelated tooutmigration does not vary in the population (i.e. ifthe_{variance V (trio (X)) . 0). This can}beverifiedforexample by

computing marginaleffects_{from binarychoice regressions on}attritionoutcomes for a sampleofindividuals who by construction donotoutmigrate, and test

if

theseeffects

aresmall. Nativesliving in thehostcountry isoneexample ofasamplenotprone to outmigration.

If attrition

which is not duetooutmigrationisbelieved toberelated to observable factors which induce sufficient variation in the attritionprocess across _individuals, nonparametric identification ofthe_{economic parametersof outmigrantsrequires some}

exclusion restriction.

Assumption A2 alo (X) = 0:10 (Xi) 2 0_{where X = (XH, X6)'}

Proposition2.2. _{IfAZ holds}andthereexists a X; given Xi such that Pr (ru = 11Xl, XS) = 0,

16 Chapter 2. Identification of Outmigration proof· Fora given X = (X;, X4):a10 (X) = alo (Xl) fromA2.Using Pr (ru = 11Xi, X;) =

0, the subsampleof immigrants

_(X;,Xi')'

_{identifies 0: 10 (Xl ) from (2.5).Given c:10 (Xl)} is identified, Pr (ru = 1IX)isidentified from_(2.6),which implies thattheweights Wo (X) and Wo (X) are bothnonparametricallyidentified. 0

Partlybecauseattritionand_{outmigration are very differentprocesses,finding}

real-isticexclusion restrictions which_{satisfy the requirementsof Proposition 2 is not very}

restrictive. _{Usually outmigration}ismodelled asa_{life-cycle event,influenced by poor}

labor market_{performance,integrationfeelings, credit rationing in the}home_country and ageatimmigration. Whether

attrition

which is not dueto_outmigrationisrelated

toallthesefactorsseemsa

priori

unlikely, given that part ofthesurvey non-response is

generally based on respondents refusal to continue working with thesurveyagencies.

2.3

Parametric

model

and

estimation

method

In this_{section, we develop and estimate}a_{parametric model which}allows us to extract

outmigrationbehaviorfrompanelattrition.The choice ofaparametricmodel is

moti-vated byourdesiretomodelselection into workand_{outmigration as}adecisionbased

on

individual

_{specific unobserved heterogeneity. We are} not aware of_{any existing}

semiparametric techniquewhich would allow us toestimatethesystemofequations presented below.

We have a measure of _{N immigrants in period} 1, where_{immigrant i remains in} the panel forTi periods. Foreach

immigrant i,

we observe in period t,whether he

pvorks pit, hismonthly labormarketearnings e(wit), and his attrition

status r;, in the

next period. The log ofthe_potentiallabormarket_earningsisassumed tobe_generated

by aloglinear earnings equation

Wit= 4,0 + Vil +

ElI (2.D

where Bareunknownparameters, Yii isan unobservedtimeinvariant individual spe-cific_componentofincomewhile_{ElIrepresents}astochasticshock. Theselabor market

earnings areonly_{observable when}an immigrant works. The

work

decision pit is as-sumed tobegenerated byalatentprocess

Section2.3.Parametricmodelandestimationmethod 17 where 0 are unknown parameters, 11,2 is an unobserved

component of work and 4

representssomestochasticshock to the

work

propensity. Participationisdetermined

by

theobservation rule pit = 1 [pi > 0]. When pit = 1, earnings wit are observed.

Both and Zl can

bethought of capturing immigrantsunobserved

ability

to generate

higherearnings and to find jobs. They can alsobethought ofasincludingunobserved

family background characteristics and preferences for workand leisure. Finally, an immigrant'sunobservableoutmigration propensity r is assumed tobedetermined by

another latent_process

ri = s;,7 + 4,3 +

2

(2.9)

where 7 are unknownparameters, _4,3 captures the

individual

specific attachment to hisnative country and 4 isa stochastic shock. The

triplet

_{{,li, 111·'13} is}assumed to be observed by theimmigrant who takes it intoaccountwhen makinghis decisions

but it is not_{observed by the econometrician. Let}

_{rM = 1 [ri > 0] be}

thedecision rule governing thetrueoutmigrationdecision

in

period t +1._{Outmigration r;;}is

unobserv-able. Inour empirical application, weassume thatAlholds6andexpresstheattrition probability as

Pr (4 = lisit) = a1O + [1 - Kio]I

Pr (r# =

lIsit)

(2.10)

Equation (2.10) is the samplecounter part ofequation (2.5).7

The earnings, work andoutmigrationoutcomes are notlikely tobeindependent of each other. This will notbe_{independent if, forexample, immigrants who find work} very easily and/or who earn a highincome are more reluctant to outmigrate. The

unobserved heterogeneity components7,1,11,2 and 4,3 canbetreated eitherasfixed

con-stants orasrandomvariables. Themain_{advantage of the fixed effect approach is that}

it doesnot requirethatincluded explanatoryvariablesbestrictlyexogenous to the un-observed heterogeneity components _{('1 , '1 ,'1 )· However, estimation}of fixedeffects

in

nonlinearmodels remains todaya sizeable_{complication, with very little guidance}

in

the choiceofmodels (see the recentreview

of

Arellanoand Honord, 2001).

A

second

drawback

of

fixed effect_{estimation is that by treating}the _{unobserved heterogeneity} componentsasfixed,cross_{equation correlations}which drive selection into work and outmigrationbasedon unobservable

individual

characteristics arenotidentified. As

the present chapterismainlyconcerned

with

selectionissues, fixedeffectestimation would limit our insights in the type ofselections present in the data. Wetherefore

in-troducethesedependenciesbyassuming that thestochastictime-invarianteffects are

covari-18 Chapter2.Identifcation of Outmigration ance matrix 4 p:.25,5, PY,35" 5,

0 = · 4 P2420,20,3

2

. 993

-where

₄

denotes_{the variances of the unobserved heterogeneity components, and p5·}

denotestheircorrelations.8 Thesecorrelationsareindicativeof whether or not

immi-grantsself-selectthemselves into work and_{into outmigration}basedontheir unobserv-able

individual

characteristics.A significantandpositivep .2indicatesthat individuals

who are more likely to work are also morelikely tohavehigherearnings,giveobserved

characteristics. p 1 has,- asimilar interpretation and isindicative of outmigration bias.

This coefficient willbenegative(positive)

if

immigrants who haveahigher

probabil-ity of

outmigratinghavebelow (above)averagemonthlyearnings. Finally, PL can be

interpretedas_{measuring outmigration bias in the}

work

_{decision and whose sign has a}

similar interpretation.

Finally, we assume that the

_{vector [el, 4,4]' is i.i.d}

normally distributed with

mean 0and covariancematrix

-2 E- -2 aw Pl,2"W pl,3gw

E= · 1 pb

1

where qi is_{the variance of}logearnings,

while

the variances of the unobserved stochas-ticshocksentering the workandoutmigrationequations are set to 1for identification

purposes. Contemporaneous correlations between the three stochastic components are

captured bythecorrelation coefficientsP ,2 Pf,3 and p6J·

To simplifythepresentation of thelikelihood function, wedividethe observable characteristics of immigranti into a set Yi = {Pit, 4, wit · Pit } il of dependent

vari-ables, a set Xi = {xit, zit, sit } ii ofexogenous variables, andavector vi = (qil, '1 , 'li)

containingunobservedtimeinvariant_{heterogeneity. Moreover, we denoteby g(·, ·. · lili)}

Section2.3. Parametricmodel andestimationmethod 19 first_stepdensityis_{given by}

fc Cy,Ixi,qi; B. 8,7, r, x1O)

Tirff

= Il / / 4 (1- rlt) (1- trio) g(p;„riA, wit;El,li) dr;,

t=i ./Qi, ./Ci,

l

J -00

CO fo

+41 f g(PA, rii, wit:ElI/i)_{0 J -00}dr;t.+-0:10 11 g (PA, r,$, wit; Elqi) dr;, dp;; drvit

The casewhereoutmigration isperfectlyobserved follows bysettinga10 equal to 0.

The sets it anditdefine thedomainofintegration over the wage and

work

spaces and

vary over timeas individualsmakedifferentchoices ineach_{period according to the}

following

table

Integrationdomainsin period t Qi: Cit

Work - [0.co)

Not Work (-co, 00) (-co,0]

Income is_{integrated out in} waveswhere individuals do not work. The _integration

domain for the

work

propensity follows from the work decision rule. In thesecond

step, theunconditionallikelihood functionis obtainedby integrating outtherandom

individual

effects over R3

f (yilxi; 4, 0,7,li, Cl,o:10) = f f (yiIX,·,1/i; 4,0,7, E, alo) h (qi; fl) dui JR3

where H denotes the trivariate normal cumulativedistribution function with mean

vector 0 and covariancematrix O.

To solvethe_{numerical integrationproblem, weapproximate theintegral by a} sim-ulated mean:asequence of r = 1,2,..., Ri.i.d.

draws 11)) = (11,1('), 42('),119(')) is taken

fromthemultivariate normal distribution H ata_givenvalue of 0.9 For each draw, the

conditional likelihood

_{function fc}

isevaluated. ThepartialMSLestimatorconsists of

replacing f bythesimulated mean

Ell, log R El, fc (Y,IXi,111'); 4,8,7,E,ir,o)11

The_resultingestimatoris inconsistentforfixed R but willbeconsistent ifRtends to

20 Chapter2.

Identifcation

ofOutmigration likelihood(Train,2003). _{Finally, we}modellabormarket_{earnings and}

work

decisions

ofGermans usingsimilarspecifications

of

_{equations (2.7) and (2.8) and estimate the}

parameters using thesimulationtechniques described above.

2.4 Data

The data used in_{this chapter}istaken fromthe_{public use file of}theGSOEPandcovers

the 1985-1999_{period. Until 1990,}the GSOEPconsisted of two samples, A and B.

Sam-ple A consistsofhouseholds

with

Germanheads

living

in formerWest-Germany Sam-ple B consists ofansampleof immigrants

living in

West-Germany coming from coun-trieswhichhad signedabilateral migrationagreement

with

Germany inthe 1950s and

1960snamelyGreece,Italy, Spain, Turkey and Yugoslavia.10 Data on speaking fluency,

integrationfeelingsof immigrants, intendedlength of stay and remittances directed to

their

family living

outside Germany were given

in

consecutive waves from1984

until

1987. Starting in 1987, thisinformationwas gathered every other yean

In

order to keep constant the timeperiod_{between observations, we have chosen to keep the}8 waves

ofthe panelwhere detailedinformation onimmigrantswas available,eachspanned

by oneyear,starting in 1985 andending in1999. _Following theliterature_measuring

the economic assimilation rate(e.g. Borjas,_1999a),werestrictour attentiontomales

between 18 and64years ofageduringthe1985-1999period. Excluded from the

sam-pleareindividuals whodied_duringtheobservation_periodand_{individuals who gave} incompleteinformation onany singlevariable enteringtheempiricalmodel in any of

the 8 waves.This leaves us withasample of1987nativeGermans and732immigrants

starting in 1985.

Theidentification_{approach presented}insection2.2relies on theinformation

con-tainedinpanelattrition. Itbecomesinstructiveto contrast theattrition pattern of our immigrantsample with thatofGermanswhose

attrition

cannotobviouslybeattributed

to_{outmigration.} Table2.1containsinformation onthenumberofindividualsobserved

along withthepercentage oftheoriginal1985samplewhoremains inagiven wave.11

41.9%ofGermans and 26.7%_{of immigrants have}beeninterviewed_{successfully in all}

the waves.Theattrition rate inagiven waveisdefined as the percentage ofindividuals not observed in_{the given wave}butobserved in the_{preceding wave. Over our}

Section 2.4. Data 21

notan issue. Assuming thatthedifferencein attritionrates is due tooutmigration, a

back of the envelopecalculation implies thatwewould_{expect theoutmigration rate in}

our sampleof immigrants to be6%_{every two years, or 3% per year,}anumberwhich would be in line

with

thosereported intheliterature(seeBorjas andBratsberg, 1996).

Of course, thiscalculationrelies on_{the assumption}thatattrition intheimmigrant pop-ulationwhich is not due_{to outmigration is ofcomparablemagnitude to that}

of

natives. We will come back to this

in

section2.5which_reportsindirect_{evidence suggesting that}

thisshouldindeed hold

in

Germany.

The top panelof Figure2.1 shows theaverage monthly gross income for work-ing immigrants andGermans over the periodcovered. In 1985, the meanincome of

Germans was 3,357 DM permonthcompared to 2,690 DMpermonth for immigrants, givinganincomeratio of1.25_{favoringGermans. The mean wage}differentialremaind

relativelysteadyuntil 1991, afterwhich, the meanincomedifferential widened even more between the twogroups toreach a_{ratio of 1.34 in 1999,}

with

Germans

receiv-inganaveragemonthly wage of 5,848DM while immigrantswere receiving 4,348 DM

per month. The divergingeconomic progressofGermans andimmigrantsafter 1991 is alsoreflected in thework_frequencies. Thebottom_{panel of figure} 2.1 shows the sample frequenciesofindividuals working inthemonthpreceding theinterview. We

can see that

until

1991, laborforceparticipation was very

similar for

both Germans

and_immigrants. After 1991, we observea_{steady decline in the}

work

_{frequencies for}

bothgroups. During that period, the percentage of immigrants

working

remained

steadily below that ofGermans. The severe drop in

work

frequencies forbothgroups

coincides withthe generaldeterioration ofthelabormarket in the regionsof former

West-Germany. Table 2.3gives the unemployment rate per yearby state. With the

exceptionofBerlin,all provinces experiencedtheirlowestunemployment rate of the

1985-1999_{period in}1991. After 1991,the_{unemployment rate}has_{progressively risen}

apart froma_{slight fall in 1999 formost provinces.}

If outmigration does occur ata systematic time inthe life-cycle, itislikely to

af-fect the age and years since immigration composition of oursample of immigrants.

Table2.2_{gives variable descriptionsand summary statistics}

for

Germans and

immi-grants for the 1985 and1999waves. We see thatbothGermansandimmigrants are, on

average, a little less than40years of age in1985while the average age ofthecohort

22 Chapter2.Identification ofOutmigration

Germansand_{immigrants leftthe panel. For}a_{given meanage,Germans have acquired}

relatively more yearsofeducation,but relatively lower labor marketexperience than

immigrants.Theaveragenumberofmonthsof labor marketexperienceof immigrants increased byalittle less than32_{months compared to}70monthsforGermans,which is

consistent with the fact thatthe_{proportion of working immigrants relative}to Germans

fell dramatically inthe 1990's.

Most immigrants migrated to Germany early in their productive lives, a fact

re-flected byan average age atimmigration

of

nearly24_years,a_figureconsistent through-out theobservation periodwhichindicates that mostmigrants wereoldenough to

au-tonomouslydecide to moveto Germany. Boththeevolutionof_yearssince_immigration and immigration yearareconsistent withthe hypothesis that outmigrationoccurs 20

years after migration (OECD, 2001). The_{average yearof immigration of our}cohort is 1969 in the 1985 wave,butraises to 1979 in the 1999 wave,indicating thattheearlier cohorts are most susceptible to havedropped out ofthepanel. Astheearlier cohorts contain themigrants withthehigher number ofyears sincemigration in 1985, it is not surprising to seethataverageyears sinceimmigrationincreases_{relatively less than the}

14 year timespan, passing from 15.75 in 1985 to 19.63 in 1999, indicatingagain that earlier cohortsare those who leftthe panel. Reported feelings onintegration in the

German society and reported speakingfluency improved over time whilehealth satis-factionseemstodeteriorate,thelatter

likely

capturinganagingeffect. Finally, 73% of

immigrantsreportedhavinga_{spouselivingoutside Gerrnany in}1985whileaslittle as 1% still do so in 1999. This result canbeinterpreted in twoways. First, spouses may have_{eventually migratedto Germanyduring the time period. Second,}

it

_{might be that}

immigrantswhosespousewas

living

abroad were more

likely

to outmigrate.

2.5 Results

and

simulations

The_regressorsincluded in_{the earnings and}work_equationsare_{education, labor}

mar-ket experience,labor market_{experience squared, selfreported}German_speaking flu-ency, and the numberofyears sinceimmigrationto Germany. These arethestandard

variables that_{have appeared in this}literature_{(Borjas, 1999a).} The_provincial

unem-ployment rate in each waveisadded in bothequationsto capture local labor market conditions. Finally, we include time fixed effects in each wave to captureremaining

ex-Section2.5. Resultsandsimulations 23

clusionrestriction in the

work

equation. Reported health satisfaction isavalid exclu-Sion restriction

if

health problems occur mostly at a time in which an

individual is

more likely to have found astablejob whosecontinuation depends onthe worker's

choices. Theoutmigrationequation includesasregressorswhether or not the wife of

immigrantslivesinGermany,theimmigrantsself reported feelingsofbelonging to the Germans society and theunemploymentrate. _{Immigrant's who arrive at}a_{young age}

presumably have the highest incentive to investin acquiring countryspecific human capital. This effectiscapturedbyincluding age atarrival

in

Germany.

Immigrants investing in their native countrypresumably havedifferentincentives

to stay in thehost country. The motives oftheseinvestments

will

affecttheir dura-tion ofstay. Immigrantsmay extendtheir stay in thehostcountry

in

ordertofurther

increase theirinvestments in their nativecountry. On the other hand,

if

immigrants

investin starting upabusiness they wish tomanage-asproposedby Dustmann and Kirchkamp (2002), high levels of investments willbeassociated

with

_shorter

migra-tion duramigra-tions. Todisentanglebothhypothesis,we include intheoutmigration equa-tionthecumulativeamountofmoneyreturned tothenative countrysince 1984 as a proxy forinvestments. Reported health satisfaction and self-reported expected length

of stayinGermany arealsoincluded, thelater captures anticipatory behavior of

mi-grants,which havebeenshown to affect the acquisitionofcountry specific skills (e.g. Dustmann, 2002b). Time dummies are_{added to capture remaining}macroeconomic fluctuations.

In ordertoseparate theimpact ofselection onearnings into a workand

outmigra-tion effect, we firstestimatedan _{earningsequation}

with

randomeffects. Oursecond

specification is abivariatemodelof labor_{market earnings and work.} We _finally es-timated the completemodelor earnings, workandoutmigration. In the latter case,

we experimented withanalternativespecification of theoutmigrationequationwhich contained education,labor _{market experience and its square, speaking fluency, and}

years sinceimmigrationas_{regressors. A log-likelihood}ratio test of the

null

_hypothesis

thatthesehumancapital variables have nojointeffectonoutmigration could not be

24 _{Chapter2. Identification ofOutmigration}

2.5.1

_Equation

results

Covariancestructure

We beginour analysis of theresults with a discussion ofthe estimates

characteriz-ingthecovariance structure ofthe unobserved componentswhichareinformative of

theselection mechanisms. Table 2.4presents estimates of the covariance structure. Focusing on the most generalmodelwhich controls for both work and outmigration selection, we findasmall_{but significant positive correlation}between theunobserved

individual_{heterogeneity of work and earnings(pY.2),}indicatingthatindividuals with higher probabilities ofworking are also morelikely tohave_{higherearnings. The}

cor-relationbetween

individual

time invariantheterogeneityof outmigration and earn-ings (p ,1 is_{\ W/} -56%while that between outmigration and work <p24 3) is 49.8%, both

.1

significant at the 1% level. Both_{correlations suggest} that individuals with a _higher

propensitytooutmigratearethose with botha_{lower probability of finding work, and} alower labor market_{earnings,which pointsto negativeoutmigrationselection. When}

comparing results withthebivariate model which does not correct

for

outmigration, we find thattheestimatedvalue of_{P ,2}remainsstable.ResultsforGermansaresimilar to that oftheimmigrant sample, withasmallbut positiveandsignificant

work

selec-tion effect

(ph)

Finally, transitoryshocksbetweenearnings and work,and shocks between workandoutmigration, areall significantly negativelycorrelated,theformer

at -34.2% and thelatter at30.4%_{while we do not find significant}correlation between thetransitoryshocks of the earnings andoutmigrationprocesses.

Outmigration

Table 2.5_{presents theparameter estimates of theoutmigration equation. We find that}

immigrantswhosewife lives with them

in

Germany haveasignificantly lower

prob-ability

of outmigrating,reflectinga_{preferencefor family unity. Immigrants}satisfied

with

theirhealthare significantlyless

likely

tooutmigrate, a finding consistent with

the sociologicalfindingsreportedinStark(1998).Intended length ofstay captures the expectations ofimmigrantsand offersdirectinformation on their remigration inten-tions.Not surprisingly, we findthatmigrants whoexpect to remain longer

in

Germany are alsolesslikelytooutmigrate. Deteriorations ofthelocal labor marketconditions, reflected

in

higherunemploymentrates, haveapositiveandsignificanteffect on the