DI

SC

US

SI

ON

P

AP

ER

S

ER

IE

S

Forschungsinstitut zur Zukunft der ArbeitInstitute for the Study of Labor

The Unknown Immigration:Incentives and Family Composition inInter-Country Adoptions to the United States

IZA DP No. 4547

November 2009

Fernando Antonio LozanoSherrie A. Kossoudji

The Unknown Immigration: Incentives

and Family Composition in Inter-Country Adoptions to the United States

Fernando Antonio Lozano Pomona College

and IZA

Sherrie A. Kossoudji

University of Michigan and IZA

Discussion Paper No. 4547 November 2009

IZA

P.O. Box 7240 53072 Bonn

Germany

Phone: +49-228-3894-0 Fax: +49-228-3894-180

E-mail: iza@iza.org

Any opinions expressed here are those of the author(s) and not those of IZA. Research published in this series may include views on policy, but the institute itself takes no institutional policy positions. The Institute for the Study of Labor (IZA) in Bonn is a local and virtual international research center and a place of communication between science, politics and business. IZA is an independent nonprofit organization supported by Deutsche Post Foundation. The center is associated with the University of Bonn and offers a stimulating research environment through its international network, workshops and conferences, data service, project support, research visits and doctoral program. IZA engages in (i) original and internationally competitive research in all fields of labor economics, (ii) development of policy concepts, and (iii) dissemination of research results and concepts to the interested public. IZA Discussion Papers often represent preliminary work and are circulated to encourage discussion. Citation of such a paper should account for its provisional character. A revised version may be available directly from the author.

IZA Discussion Paper No. 4547 November 2009

ABSTRACT

The Unknown Immigration: Incentives and Family Composition in Inter-Country Adoptions to the United States Children adopted from abroad are an immigrant group about which little is known. According to the U.S. Census more than one and a half million children living in the U.S. are adopted, with fifteen percent of them born abroad. In fact more than twenty thousand adopted orphans from abroad enter the country each year. The families who adopt these orphans are mostly white, wealthy, and well educated (Kossoudji, 2008). What are the characteristics of children who are adopted from abroad and what incentives drive families to adopt them? In this paper we use the 2000 census to illuminate the landscape of international adoption. We address three issues: 1) How do the demographic characteristics of the children adopted from abroad change as other countries open and shut the door to inter-country adoptions, changing the supply of available children? 2) U.S. born parents and foreign-born parents may have different incentives to adopt. How are these incentives reflected in the characteristics of the children they adopt? 3) What explains differences in the estimates of foreign-born adopted children in the Census and the number of visas granted by the State Department? JEL Classification: J13, J12 Keywords: immigrant children, international adoption Corresponding author: Fernando A. Lozano Department of Economics Pomona College 425 N. College Ave Claremont, CA 91711 USA E-mail: fernando.lozano@pomona.edu

1

I.Introduction

AccordingtotheU.S.Censusmorethanoneandahalfmillionchildrenunder

theageofeighteenlivingintheU.S.areadopted.Fifteenpercentofthesechildren

werebornabroad.Whiletheadoptionofforeign‐bornchildrenwasarelatively

isolatedoccurrenceearlyinthecentury,thenumberofforeign‐bornadopted

childrenarrivingintheUnitedStatesonadoptedorphanvisasrosefromjustover

8,000in1989(thefirstyearofnationaldatacollection)tomorethan18,000in

2000,theyearofthecensus.Theinflowoftheseforeign‐bornchildrenhasreshaped

thecompositionoftheaverageAmericanfamily,especiallyasadoptionhasbeen

usedasameansforfamilyformationamongparentsunabletohavetheirown

children(Bachrach,1983).Inthispaperweexplorethechangingsupplyofchildren

fromdifferentcountries,thecharacteristicsofchildrenadoptedfromabroad,and

howthesecharacteristicsvaryacrossparentswithdifferentcharacteristics.

Unsurprisingly,economistsandothersocialscientistshavepaidattentionto

theoutcomesofchildrenadoptedfromabroad(see,forexample,Sacerdote,2007;

Plug,2004).Yetthisliteraturedoesnotdiscussanysupplyissues.Inthisliteratureit

isassumedthatallparentswishingtoadoptfromabroadfacesimilarcostsand

incentives;andwillpursueacertainsetofcharacteristicsinchildren.Implicitly,

adoptingparentssearchforcharacteristicsthatwillsuccessfullyleadto“family

formation”.Forexample,thesecharacteristicsmaybeageorthegenderofthe

adoptedchild.Therearetworeasonswhyveryyoungchildrenaremoreappealing

amongadoptingparents:first,whenparentsadopttheywanttomaximizethe

lifetimeutilityderivedfromparenting,andmoretimeasaparentincreasesutility.

2

Second,childrenadoptedatanolderagemayhavemoreexposuretotheeffectsof

anadversebackground.Brooksetal.(2002)reportthat85%ofprospective

adoptingparentsarewillingtoadoptachildtwoyearsoryounger,whileonly45%

arewillingtoadoptachildoversixyearsold.Butparentsmayfaceverydifferent

incentivesandcostswhenadoptingfromabroad,andthesedifferenceswillyield

systematicdifferencesintheadoptionprocessandinthecharacteristicsofthe

childrenadopted.Parentsmaychoosetoadoptforhumanitarianreasons(child

centeredreasons),forexample,anincentivethatisnotdiscussedinthisliterature,

andmayfindadoptingolderchildrenmoresatisfactory.

Wefindthatthereareverycleardifferencesbetweenadoptingparentsborn

intheUnitedStatesandthosebornabroadintheiradoptingchoices.Whilethe

formertendtoadoptbabiesandinfants,thelattertendtoadoptolderchildren.In

addition,foreign‐bornparentswhoadoptchildrenfromtheirownbirthcountry

adoptchildrenwhoare,onaverage,olderthanthoseadoptedbyotherforeign‐born

parents.Theseresultscanbepuzzlingifweonlyconsiderthefamilyformation

incentivesoutlinedabove.OurevidencesuggeststhatU.S.bornparentsdoappearto

respondtotheincentivesthatleadtofamily‐formationadoption.Incontrast,

foreign‐bornadoptingparentsmayhavedifferentincentivestoadoptchildren,beit

thattheseparentsfacesocialpressuresfromtheircommunitiesoforiginorhavean

altruisticutilityfunction,whichismaximizedbyadoptingolderchildrenfromtheir

communitiesoforigin.Thecontributionsofthesepaperarethree‐fold:first,we

documentthatparentalcharacteristicsareimportantdeterminantsofadopted

children’scharacteristics,especiallydifferentparentcharacteristicsthatare

3

associatedwithdifferentchild’sageatadoption;second,weexplorehowthese

characteristicsvaryacrosstimebetween1983and2000;thirdweconsidersupply

issuesandthevalidityofusingthecensustotalkaboutforeignbornadoptionsby

comparingtheestimatesofthenumberofadoptedchildrenusingthe2000

DecennialCensus,withthenumberofvisasgrantedbytheUSDepartmentofState.

II.The2000CensusandAdoptedChildren

Weusethe5%sampleofthe2000U.S.Censusforthedatainthispaper.The

2000censusasked,fortheveryfirsttime,therelationshipstatusoftheheadofthe

householdtoanyadoptedchildreninthehousehold.Beforethe2000census,

biologicalchildrenwereindistinguishablefromadoptedchildren.Werestrictour

sampletoadoptedchildrenundertheageof18.Inthe2000censustherewere59.8

millionbiologicalchildren,1.6millionadoptedchildren,and3.3millionstep‐

childrenundertheageofeighteenlivingwithhouseholdersintheUnitedStates.To

investigateonlyforeign‐bornadoptees,weuseadditionalinformationfromthe

census.Thecensusprovidesinformationontheplaceofbirthforallindividualsand,

forthosebornabroad,theyeartheimmigrantarrivedintheUnitedStates.The

actualdateofadoptionisnotenumeratedinthecensus.Usingtheimmigration

information,weinfertheage,andyear,atwhichachildbornabroadwasadopted

byassumingthattheyearthechildenteredtheUnitedStatesistheyearofadoption.

Wedonotknowifaforeignbornchildwasactuallyadoptedsometimebefore

immigratingtotheUnitedStates,sotheyearofadoptionweestimatemaybelater

thantheactualadoptiontookplace.Inmanycasesthisisonlyaminorproblemas

4

manychildrenadoptedfromabroadareadoptedasinfantsortoddlers.Becausewe

focusoninter‐countryadoptionsweexcludefromthesamplealladoptedchildren

bornabroadwhoseparentswerebornabroadaswellandwhoseyearof

immigrationisthesameaseitherparent.Insomeanalyses,wecomparetheageof

adoptionindifferentyears.Sinceolderchildrenhaveagedoutofthehousehold,and

childrenadoptedatolderagesinearlieryearswillhavealreadyleftthehousehold,

werestricttheyearsofanalysisinthosecases.Whilethecensusisnottheoptimal

datasettoexaminechangesinadoptionovertimebecauseofthisproblem,it

affordsusafirstglimpseofchangingadoptionintheUnitedStatesduringacritical

periodofitsgrowth.

II.TheDemographyofAdoption

Figure1showsthenumberofadoptedforeign‐bornchildrenbyyearofbirth,

andthenumberofadoptedforeign‐bornchildrenbyyearofimmigration.We

includeboththeyearofbirthandtheyearofimmigrationinFigure1becausewedo

notknowtheexacttimeofadoption.Thenumberofadoptedforeign‐bornchildren

whoimmigrateeachyearincreasesthroughouttheentireperiodandskyrockets

after1993.Fortheearlieryears,thelownumbersmaybeanartifactofthedata,as

childrenwhowereadoptedintheearlieryearsaremuchmorelikelytobeoverthe

ageof18and/ornotlivinginthehouseholdduringthe2000census.Thus,the

numberofforeign‐bornadoptionsrecordedintheU.S.2000Censusisan

underestimateoftheactualnumberofadoptionsduringeachrecordedyear.

However,thegrowthofadoptionsinthe1990sisarealtrend.ThenumberofU.S.

5

visasforadoptedchildrenrosethroughoutthedecadebecausechangesinthe

adoptionlawsinChina,Guatemala,andelsewhere.Thispositivesupplyshock

allowedAmericanfamiliestofindthedoorsfortheadoptionofyoungchildrenin

manycountriesopen.

Insummary,Figure1showsthattheagedistributionforchildrenborn

abroadismuchmoreuniformthanthetimeofmigration,indicatingthat(ofthe

childreninthehouseholdatthetimeofthecensus)childrenwereadoptedat

differentages.Eventhoughtheadoptionofbabiesfromabroadisthestereotype,

olderchildrenareadoptedaswell.

III.ForeignBornAdoptionandAdoptiveParents’Birthplace

Childrenwhoareadoptedfromabroadhavealwaysbeeninaunique

citizenshipcategoryandgrantedcitizenshipmorereadilythanmostother

immigrants.PracticewascodifiedrecentlywiththeChildCitizenshipActof2000,

whichmeansthatcitizenswhowishtoadoptabroadhavetherighttopetitionfor

theimmediateimmigrationofachildadoptedabroad.Iftheadoptionisfinalized

abroad,thechildbecomesacitizenuponentrytotheUnitedStatesusingtheIR3

visa.IftheadoptionisfinalizedintheUnitedStates,thechildenterstheUnited

StatesonanIR4visaandbecomesacitizenautomaticallywhenthecourtissuesthe

finaladoptiondecree.Legalresidents,however,mustpetitionforimmigration

underfamilypreferences,whichcouldtakeyears,althoughthereareexceptions:1

1 “The U.S. Department of State has clarified that U.S. permanent residents may petition for their adopted alien child to immigrate as a qualified derivative only if they completed the adoption prior to their admission as lawful permanent residents. In other words, only children who were adopted before their

6

Adoptedchildrenoflegalresidentscanpetitionforcitizenshipaftertheageof

eighteen.

Potentialparentsmayhavedifferentincentivestoadopt.Mosteconomists

thinkofthetimespentwithone’sadoptedchildrenasanormalgood,andsothe

moreadvancedagetheadoptedchildis,thelowerwillbethepresentvalueofthe

utilityderivedfromadopting.Asaresult,parentswhoadopthaveincentivesto

adoptchildrenatasyounganageaspossible.Theprevalenceofforeign‐bornbabies

adoptedbyparentsintheUnitedStatesspeaksoftheseincentives.Parentswhoare

bornintheUnitedStatesarelikelytofitthemodeloftheinfertilecoupleorthe

coupledrivenbycharityincentivestoadoptayoungchildfromabroad.Butthe

incentivesforadoptionmayrestondifferentfactorsforforeign‐bornparentsinthe

UnitedStates.ParentswhowerebornabroadandimmigratedtotheUnitedStates

mayalsohaveincentivestoadoptbasedonfamilyorcommunityobligationsinthe

homecountryoronaltruisticmotivesrelatedtothehomecountry.These

obligationsormotivesloosenthetiebetweentheageofthechildandutilityor

benefitsfromtimespentwiththechild.

Asaresult,therearetwoprincipalsourcesofadoptionofchildrenborn

abroadandthecensusdatashowhowthosetwokindsofadoptionreflectthe

characteristicsoftheadultswhoadoptthem.Morethanthree‐quartersofthe

childrenadoptedfromabroadareadoptedbyparentswhowerebothborninthe

UnitedStates.Parentsborninothercountriesalsoadoptchildrenfromabroadand

parent became a Green Card holder and who satisfy the three requirements for adoption: (i.e., (1.) adoption before the age of 16; (2) two years custody by their adoptive parent; and (3) two years of residence with the parent) may immigrate into the U.S. together with their parent as a derivative.” Again, these children are excluded from the sample.

7

householdswhereatleastoneparentisforeign‐bornareresponsibleforone‐

quarterofallchildrenadoptedfromabroad.Thesetwokindsofhouseholds,

however,haveverydifferentadoptionpatterns.Figure2showstheageofbirthand

theyearofimmigrationforchildrenadoptedbytwoU.S.bornparents(Panel1)and

thoseadoptedbyleastoneparentwhoisforeignborn(Panel2).Singleparents

withoutpartnersareincludedineachpanelbycountryofbirth.Theyearofbirth

andtheyearofimmigrationareverysimilarinPanel1whentheparentsareboth

U.S.born.ChildrenadoptedbytwoU.S.bornparentsareadoptedatanaverageage

ofthreeyearsold.Immigrationnumbersrisealongwithbirthdate.Largenumbers

ofchildrenborninayearareassociatedwithlargenumberofchildrenimmigrating

inayearwithapossiblesmalllag.Thesmalldipinthenumberofbirthsintheyears

justbeforethecensusyearindicatesthatbabiesonlyfromthosebirthyearshave

alreadybeenadopted.InthesecondpanelofFigure2,however,birthyearand

immigrationyearmoveinoppositedirections.Thenumberofchildrenimmigrating

eachyearrisesbutthenumberbornineachyearfallsovertimeforthesechildren.

Thereisahighproportionofadoptedchildreninthe2000censuswhowerebornin

the1980s.Butthehighestproportionofimmigrantchildrencameinthelate1990s.

Childrenadoptedbyforeign‐bornparentsareolder.Infact,theaverageageof

foreign‐bornchildrenatadoptionwhenatleastoneparentisforeignbornismore

thanfiveandahalfyearsold.

Parentswhoarebornabroadmayhaveincentivestoadoptthatare

systematicallydifferentfromthoseofparentsbornintheUnitedStates.They,too,

mayormaynothavefertilityproblems.Theymayadoptachildbecauseofthe

8

deathofarelative.Inaddition,theymayhavefamilyorcommunityobligationsor

besubjecttopressurebypeoplefromhometoadoptachildtogivehimorhera

betterlife.Theymaysimplyfeelmorealtruistictowardchildrenfromtheirhome

countryandadoptforthesamehumanitarianreasonsbutwithamoretargeted

focus.Asaresult,two‐thirdsofthechildrenadoptedbyforeign‐bornparentscome

fromtheirhomecountry.Thenextfigure,Figure3,showsthesametrendsasFigure

2butforforeign‐bornparentswhoadoptfromthehomecountryandforeign‐born

parentswhoadoptfromadifferentcountry.Whilebothgroupsofparentsadopt

childrenwhoareolder,onaverage,thanthoseadoptedbyU.S.bornparents,those

adoptedfromthehomecountryaremucholderthanthoseadoptedfromadifferent

country.Theaverageageofadoptionissixandahalfyearsoldfortheformer,but

lessthanfouryearsoldforthelatter.

Figure4documentsthiscrucialdifferenceintheageatadoptionofchildren

adoptedbyforeign‐bornparentsandU.S.bornparents.Firstnoticethat,justasin

theothergraphs,anartifactofthedataisthatthelevelsofadoptionaretruncated

forhigherages.Buttheshapeoftheadoptionpatternsstillshowsthroughinthis

graph.Thedifferenceintheagesofadoptionforforeign‐bornadoptedchildrenare

starkwhencomparingparentswhenbothbornintheUnitedStatesandparents

whenneitherisbornintheUnitedStates.WhenbothparentsareU.S.born,nearly

one‐halfofalltheadoptedchildrenarebabiesatadoption.Another18percentare

ageoneand10percentareagetwo.Over84percentofchildrenareadoptedatage

fiveorbelow.Afterthat,thereisalowbutsimilarpercentageacrosstheages.

FamilieswithatleastoneparentU.S.bornhavekidswithsimilarcharacteristics,but

9

withaslightlylowerpercentageofinfantsadopted.Incontrast,whenbothparents

areforeign‐born,only14percentoftheadopteesareinfantsand10percentare

adoptedatageone,andtherearelargenumbersofchildrenadoptedatallages.

Morethan50percentofchildrenareadoptedatagesixorabove.Insummary,

Figures2to4showstarkdifferencesbetweentheagecharacteristicsofchildren

adoptedbyU.S.bornparentsandchildrenadoptedbyparentsbornabroadlivingin

theUnitedStates.Asaresult,inter‐countryadoption,itself,needstobeconsidered

separatelydependingonthebirthplaceandincentivesoftheparentswhoadopt.

IV.CountryPolicyandthesupplyofchildren—Countryofbirth

Thecharacteristicsofcohortsofadoptedchildrenpresentapictureof

geopoliticsandcountrydisaster.LargescaleadoptionbeganafterWWIIas

AmericansadoptedchildrenfromwartornEuropeandJapan,andKoreanadoption

followedtheKoreanWar.Thenewestwaveofinter‐countryadoption,however,

whilearelativelysmallcomponentofoverallimmigration,issignificantbecause

countryspecificpoliciescanhavesuchalargeimpactonthesourcecountriesof

adoptedorphans.Theseadoptedbabiesandyoungchildrencomeinwavesfrom

specificcountriesinresponsetoeventsorpolicychangesinthosecountries.Figure

5AshowstheStateDepartment’svisaentrantsforallyearsbetween1989and

2008.2Figure5Bshowsforeignbornadoptedchildreninthe2000Censuswho

arrivedintheUnitedStatesbetween1984and1999.

2 The ten countries that accounted for the largest number of adopted orphan visas during the census period (1989-1999) are individually documented in Figure 5A. This is not the same set of countries that accounted

10

Often,singulareventsorpolicieschangethecharacterofinter‐country

adoption.Asnoted,KoreahasbeenasourceofadoptedorphanssincetheKorean

War.Koreawasstillresponsiblefor44percentofallforeign‐bornadopteesin

1989.Thatyear,theKoreangovernment,respondingtodemocratizationand

negativepublicityaboutthelargenumbersofchildrenadoptedandsentabroad,

vowedtoreduce,andeventuallyeliminate,adoptionsofKoreansbypeoplefrom

othercountries.Thispolicywasnotcompletelysuccessful,butasFigure5Ashows,

adoptionsfromKoreawentfrom44percentofalladoptionsto12percentofall

adoptionsinjusttenyears.Overthetwenty‐yearperiod,asthenumberofadoptions

fromallcountriesincreased,boththepercentageandthenumberofadoptionsfrom

Koreadecreased.Lessthanone‐thirdasmanychildrenwereadoptedfromKoreain

2008thantwentyyearsbeforeeventhoughoveralladoptionmorethandoubled.

Similarly,theoverthrowofNicolaeCeausescuofRomaniain1991wasaccompanied

bypublicityabout“thousandsofchildren”inunderstaffedorphanagesbecauseof

populationexpansionpolicy.3U.S.adoptionsfromRomania,numberingonlya

coupleofhundredbefore1991,roseto2,594during1991,makingRomanians31

percentofforeign‐bornadoptedorphansin1991.Immediately,criesofwidespread

corruptionandthediscoveryofproblemsamongRomanianchildrenwhowere

adoptedledtoatemporarymoratoriumoninter‐countryadoptionsfromRomania

attheendof1991.Asaresult,almostnoRomanianchildrenwereadoptedin1992

or1993.Attemptstopromotebothinter‐countryadoptionandlocalsocialservices

for the largest number of visas during the entire period up to 2008 as country policy increased or decreased the number of children available to be adopted by parents in the United States. 3 http://www.adoptioninstitute.org/FactOverview/international.html

11

ledtoincreasesinadoptionfromRomaniauntilaharshermoratoriumwasput

throughafterthe2000censusandadoptionsfromRomaniawerevirtuallyhalted.

Romanianadoptionwasreplacedbyadoptionfromothercountries.Although

therewasalmostnoadoptionfromRussiaortheSovietsatellitesbefore1992,the

break‐upoftheSovietUnionwasresponsibleforatidalwaveofadoptionfromthe

formersatellitesandRussia,itself.Asorphanagesfilledtocapacity,Russiaopened

itsdoorstointernationaladoption.SodidKazackstan,Ukraine,andBulgariaand

othersovietsatellites.AllshowagrowingstreamofchildrenadoptedintheUnited

States.Bythelate1990s,overone‐thirdofallforeign‐bornadoptedorphanscome

fromtheformerSovietUnion.

Expansionarypolicywithincountrieswasahaphazardprocess,probably

respondingtotheemergenceofnascentadoptionnetworksandlocalstateneeds

andissues.TherewasessentiallynoadoptionfrommainlandChinauntil1992,

whentheChinesegovernmentpassedTheAdoptionLawofthePeople’sRepublicof

Chinathatallowedinternationaladoptions.Chinahadanabundanceofadoptable

girls,probablyasaresultoftheone‐childpolicyandfamilieswithinChinapreferred

boysandsotherewaslittleadoptiondemandforgirls.AdoptionfromChinajumped

from0.7percentofthetotalin1991to29.4percentin1996andChinaremainsthe

largestcountrygroupingofforeign‐bornadoptions.Guatemalaisoneofthefew

countriesthatallowssingleparentstoadoptandadoptionfromGuatemalagrew

throughouttheperiod.AlthoughGuatemalawasarelativelylowpercentageofvisas

beforethe2000census,Guatemalanorphansplayamuchlargerroleafterthe

census.

12

Importantly,Figure5A,fromStateDepartmentVisas,andFigure5B,fromthe

U.S.Census,showssimilarlyhighpercentagesofchildrencomingfromthreemajor

sourcecountries:Korea,ChinaandRussia.Thereisasignificantlyhigherpercentage

ofKoreanchildreninthecensusdata.Thesefiguresdodifferespecially,however,in

thenumberofadoptedchildrenfromthreecountriesthataretraditionallabor

exportingeconomiestotheUnitedStates:Mexico,Guatemala,andthePhilippines.

Wenowknowthatthebirthplaceoftheparentsmattersforthebirthcountry

oftheadoptedchildaswell.Tofurtherexplorethesedifferences,Figure5Cand

Figure5Ddividethesampleagainbynativityoftheparents,whereFigure5C

documentshouseholdswithbothparentsU.S.bornandFigure5Ddocumentsthose

childrenadoptedbyforeign‐bornparents.WhenbothparentsareU.S.born,the

sourcecountrydistributionfollowsthehistoryofinter‐countryadoptionand

countrysupply.Thisfigureclearlyidentifiesothercountries’policies.Betweentwo‐

thirdsand89percentofchildrenarefromthetoptensendingcountriesandmostof

theadoptionscomefromKorea,China,andRussia.Butwhenparentsareforeign‐

born,thesourcecountrydistributionappearsmorelikethesourcecountry

distributionforimmigrantstotheUnitedStates.ThecountrieshighlightedinFigure

5DaredifferentfromFigure5C;whileMexico,thePhilippines,ElSalvadorand

DominicanRepublicarebarelyfeaturedinFigure5C,thesecountriesarefeatured

predominantlyinFigure5D.

AsaresultofthedifferentincentivesforU.S.bornandforeign‐bornparents

andthewaysthatchildrentheyadoptmaygainentrytotheUnitedStates,visa

statisticsfromtheStateDepartmentmayunderestimatethenumberofforeign‐born

13

adoptedchildrenenteringthecountryeachyear.Whiletheremayberecallerrorin

thecensusabouttheexacttimingofadoptionandimmigrationtotheUnitedStates,

thesecensusdatarepresentthebestestimatesofadoptiontrends.InFigure6,we

comparetheadoptedforeign‐bornorphanvisasforeachyearwiththenumberof

adoptionsreportedtohaveimmigratedeachyearbyfamiliesreportinginthe2000

census.Eventhoughthecensusnumbersareunderestimatesbecauseofagingout,

therearemany,manymoreadoptionsinthecensusthaninthevisafigures.

Althoughthetwoseriesstartoutsimilarly(thevisastatisticsareonlyavailablefrom

1989)thereisadivergenceintheearly1990sthatgrowssubstantiallyuntilthere

arethousandsmorechildrenimmigratingineachyearaccordingtothecensusdata

thanthevisadata.Giventhatthesearenotforeign‐bornchildrenwhoimmigrated

withtheiradoptiveparentsandthatthecensusdataareunderestimates,evenin

theseyearsclosetothecensus,thelargedifferencesinthesenumbersindicatethat

westilldonothavegoodestimatesofthenumberofforeign‐bornadoptedchildren

whoimmigrateeachyear.

V.Parents’CharacteristicsandtheCharacteristicsofAdoptedChildren

Nextweexplorehowthecharacteristicsoftheadoptingparentscorrelateto

theagewhenthechildwasadopted.Table1presentsthedemographic

characteristicsofadoptingparents.Wedividethedataintothethreedifferent

categoriesreflectedinFigure1toFigure4.Thefirstcolumnpresentssamplemeans

forparentswhowerebornintheU.S.,thesecondcolumnpresentsthesample

meansofparentsinhouseholdswhereatleastoneparentwasbornabroadandwho

14

adoptedintheirbirthcountry,andthethirdcolumnpresentsthesamplemeansof

parentsbornabroad(atleastone)andwhoadoptedfromadifferentcountrythan

theirbirthcountry.

Thefirstrowpresentstheestimatedageofthemotherandfatherwhenthe

childwasadopted.Thereareonlysmalldifferencesintheestimatedagecategories:

asforeign‐bornparentsadoptinginadifferentcountrytotheirplaceofbirthare,on

average,theeldestatthetimeofadoption.Theyarefollowedbyparentsborninthe

U.S.,andtrailedbyparentsbornabroadwhoadoptchildrenintheirbirthcountry.

Thesecondandthirdrowspresenttheproportionofmothersandfathers

respectivelywhograduatedfromcollege.Foreign‐bornparentswhoadoptintheir

countryofbirthplacearehalfaslikelytograduatefromcollegethanotherparents

adoptingabroad.Morethanone‐halfofthemothersandfatherswhoareU.S.born

orwhodonotadoptfromtheircountryoforiginarecollegegraduates.

Table1alsoincludessomelabormarketinformationofadoptingparents.

Thefourthrowdocumentsthehousehold’s1999annualincome;againforeign‐born

parentswhoadoptintheirowncountryearnlessthan65percentofwhatother

adoptingparentsearn.Sinceadoptionsoccurredatdifferenttimes,andweonlysee

earningsoutcomesin1999,thisincomedoesnotrepresentagoodcomparisonof

incomedifferencesifforeign‐bornparentsadoptedchildrenatadifferenttimeorat

adifferentagethanUSbornparents.Tohelpassessincomedifferences,the

estimatesinthefifthrowcontrolforagedifferencesbyshowingtheestimateofthe

15

householdearning’spercentilewithin7agegroups4,thisisanumberbetween0

(lowestearners)and100(highestearners)andrepresentsthehouseholdposition

intheincomedistributionofthehead’sagegroup.Again,theaverageforeignborn

parents’earningsadoptingintheirhomecountryfallinthe55thpercentile,while

theaverageforeignbornparentsearningsadoptinginadifferentcountryareatthe

70percentile.TheaverageU.S.bornparents’earningsareinthe73rdpercentile.

Parentswhoadoptchildrenfromtheirhomecountryearnlessthanotherparents

whoadoptfromabroad.

Table1alsoincludesinformationonfamilycomposition.Thesixthand

seventhrowsdocumenttheproportionofadoptinghouseholdswithasingleparent

(theseincludedivorcedandseparatedparents).Therearenotmanysignificant

differencesacrossparentgroups:typicallyaboutoneinsixadoptingfamilies

consistsofasingleparent.Finally,foreign‐bornparentsaremorelikelythanU.S.

bornparentstohavetheirownbiologicalchildrenatthetimeofadoption,About20

percentofthoseadoptingfromtheirhomecountryand19percentofthoseadopting

fromadifferentcountryhavetheirownbiologicalchildreninthehousehold.Only

16percentofU.S.bornparentsdo.Butlookingatallchildreninthehousehold

reversesthatcomparison.U.S.bornparentsandforeign‐bornparentswhoadopt

fromadifferentcountryaremorelikelytohavechildreninthehouseholdthan

foreignbornparentswhoadoptfromthehomecountry.

4Theagegroupsforadoptingparentsare:lessthan25,25‐29,30‐34,35‐39,40‐44,45‐49and49+.Theresultsaresimilarifinsteadweuseearningsadjustedforagewhenthechildwasadopted.

16

Table2documentsthecharacteristicsofallforeign‐bornadoptedchildren

usingthesamecolumngroupsasinTable1.Thefirstrowpresentstheaverage

child’sagewhenadopted.ChildrenadoptedbyU.S.bornparents,areadoptedon

averageatagethree.Childrenadoptedbyparentswhowereborninthesame

countryareadoptedonaverageat6.5yearsofage.Childrenadoptedbyforeign‐

bornparentsadoptedinadifferentcountrythantheparents’birthplaceareadopted

onaverageat3.7yearsofage.Therearealsodifferencesonthegenderofthe

adoptedchild;parentsthatadoptfromtheirhomecountryareslightlylesslikelyto

adoptgirlsthanallotherparents.Foreign‐bornparentwhoadoptfromtheirhome

countryaremuchmorelikelytoadoptachildwithHispanicethnicity:almostone

halfofthoseparentsadoptaHispanicchild.Incontrast,only20percentofother

parentsadoptanHispanicchild.Similarly,parentsadoptingintheirbirthcountry

aremuchmorelikelytoadoptachildwhoisblack.Whiletherearemany

characteristicsofadoptiveparentsthataresimilar,thesamplemeansabove

highlightthattheycomefromdifferentincomeandeducationalexperiencesand

theyadoptchildrenwithdifferentcharacteristics.

Asweargueinthispaper,animportantcharacteristicofadoptedchildren,

andanimportantdifferenceoftheadoptionforparentswhoarebornintheUnited

Statesorabroadistheageofthechildwhenheorsheisadopted.What

demographiccharacteristicsdeterminetheageatwhichachildisadopted?While

wecannotspeaktooverallsupplyofchildrenlookingtobeadoptedbyage,wecan

lookattheageofadoptedchildrenandthewaythatparents’characteristics

correlatetotheageofadoption.Tounderstandhowparents’characteristics

17

correlatewiththeagewhenthechildwasadoptedweestimatethefollowinglinear

regressionmodel:

€

Ait = θ1Xit ×USit + θ2Xit × SCit + θ3Xit ×DCit + C'β +Y 'γ + εit (1)

whereAistheestimatedageofadoptionforthechild,Xitisavectorofparents

demographiccharacteristicsandincludesthemother’sagewhenadopted,whether

themotherisacollegegraduate,whetherthemotherisasingleparent,theage

adjustedhouseholdincomepercentileandwhetherotherchildrenwerepresentin

thehouseholdatthetimeoftheadoption5.Thesecharacteristicsareinteractedwith

threedifferentindicatorvariables:USit,SCitandDCitwhichstandforU.S.born

parents,forforeignbornparentsadoptinginthesamecountryastheywereborn,

andforforeignbornparentsadoptinginadifferentcountryrespectively.θ1,θ2and

θ3arevectorsofparametersassociatedwitheachinteractiontermbetweenthe

parents’characteristicsandtheparents’birthplace.ThevectorCcontainsfixed

effectsforKorea,China,Russia,Mexico,India,Colombia,Guatemala,Romaniaand

Philippines,andthevectorYrepresentsyearfixedeffects(neitherofwhichis

interactedwithparentcharacteristics).Whenestimatingthisregressionweonly

includeobservationsofchildrenadoptedin1990orlater,aspotentiallysomeofthe

childrenadoptedbeforearealreadyoutofthehouseholdbytheyear2000;the

resultsarerobusttowhetherweincludechildrenadoptedbefore1990ornot.

Theestimatesforthecoefficientsinequation1arepresentedinTable3.

Recallthatinthisequationeachparentcharacteristicisinteractedwiththe

dummiesUS,SCandDC.Thefirstcolumnpresentstheestimatesofthecoefficients

5Forhouseholdswithoutmothers,theinformationforthefatherisused.

18

fortheinteractionsbetweeneachcharacteristicandUS.Thesecondcolumn

presentstheestimatesofthecoefficientsfortheinteractionswithSC,andfinallythe

thirdcolumnpresentstheestimatesfortheinteractionswithDC.Thebottompanel

presentstheestimatesforeachcountryfixedeffect6.TheomittedcategoryisallU.S.

bornparentsadoptingabroad,henceallestimatesarerelativetothisgroup.The

estimatesinthefirstrowconfirmtheresultsfromFigures1‐4,thatis,controlling

formother’sdemographiccharacteristicsforeign‐bornparentsadoptchildrenwho

areoldenthanchildrenadoptedbyU.S.bornparents.Whentheparentsadoptin

theirhomecountry,childrenareconditionalonotherdemographiccharacteristics,

almost8yearsolderattheageofadoptionthanchildrenadoptedbyU.S.born

parents–recallthatthesampleincludesonlychildrenwhomigratedatadifferent

timethaneitherparent.ConditionalonotherdemographiccharacteristicsWhen

foreign‐bornparentsadoptinadifferentcountry,theirchildrenareonaverage,

almost4yearsolderatageofadoptionthanchildrenadoptedbyU.S.bornparents.

Oldermotherstendtoadoptolderchildren,yettherelationshipbetweenthe

ageofthemotherandtheageofthechildisalmostthreetimesstrongerformothers

whoareU.S.bornthanforforeign‐bornmotherswhoadoptintheirbirthcountry.

Thatis,thelinkbetweenmother’sageandchild’sage,whilesignificantforforeign‐

bornparents,isquitesmallforthosewhoadoptfromtheirbirthcountry.Education

isinverselyrelatedtochild’sage:thatismotherswithacollegedegreearemore

likelytoadoptyoungerchildren.Therelationshipbetweenmother’seducationand

child’sageatadoptionisstrongestfortheforeign‐bornwhoadoptintheirbirth6Weomittheestimatesforyearfixedeffects,theseareavailablefromtheauthorsuponrequest.

19

country.Finally,familycompositionseemstohavenorelationshiptochild’sage.

Thereisnostatisticallysignificantassociationbetweenthemother’smarriedstatus

andthechildren’sagewhenadopted;similarly,household’sincomeisonly

negativelyassociatedwithchildren’sageatadoptioninthecaseofforeignborn

parentsnotadoptingintheircountryofbirth.Finally,ifotherchildrenarepresent

inthehousehold,youngerchildrentendtobeadopted.Thiscorrelationisstrongfor

allthreeparentalinteractions.Notethatthecountries’fixedeffectsreflectthe

estimatesinFigure5A‐D,wherechildrenadoptedfromChina,Korea,Russia,India

andGuatemalatendtobeyoungerwhenadopted,whilechildrenadoptedfrom

MexicoandthePhilippinestendtobeelder.

TheestimatesinTable3implicitlyassumethattheassociationbetween

parentcharacteristicsXandthechild’sageofadoptionisconstantacrossallage

groups.Thereisnoreasontothinkthisisavalidassumption,ifparentshavea

strongpreferenceforinfantchildren.Asanalternative,weestimatethefollowing

logitregression

€

Iit = θ1Xit ×USit + θ2Xit × SCit + θ3Xit ×DCit + C'β +Y 'γ + εit (2)

whereIitisavariableindicatingtheprobabilitythatachildisnotaninfant,defined

inthiscaseasachildwhoisthreeorolder,attheageofadoption.Iitisnotobserved,

insteadweuseI’itwhichtakesavalueofoneifthechildwasadoptedafterherthird

birthday,zerootherwise.Estimatesforequation(2)arepresentedinTable4.

TheestimatessuggestastorythatissimilartotheonerevealedinTable3

withtwoimportantexceptions,childrenadoptedbysingleparentsfromthesame

countryofbirthare20percentagepointsmorelikelytobethreeyearsofageor

20

older.Inaddition,thehousehold’sageadjustedincomeisnegativelyassociatedwith

theprobabilityofadoptingachildagethreeorolder.This,togetherwiththe

negativeestimatesforthecoefficientofmotherhavingacollegedegree,is

suggestivethatfamilieswithmoreresourcesanddesirablecharacteristicsaremore

likelytoadoptinfants,consistentwiththefamilyformationhypothesis.Inaddition,

oncethatwecontrolforthesecharacteristics,childrenadoptedbyparentsbornin

thesamecountryare70percentagepointsmorelikelytobethreeorolder,children

adoptedbyforeignbornparentsinadifferentcountryare52percentagepoints

morelikelytobethreeorolder.Insummary,notonlydodifferencesintheageof

adoptionpersistonceonecontrolsforparentscharacteristicsandcountryofbirth,

butactuallythesedifferencesmagnifywhenwecompareveryyoungchildrentoall

others,againsuggestingthatparentsbornintheUnitedStatesandforeign‐born

parentsadoptverydifferentchildren.Thisisespeciallysowhentheparent’sadopt

childrenintheirbirthcountry.

VI.Summary

Theadoptionofforeign‐bornchildrenbyparentslivingintheUnitedStateshas

morethantripledduringthelastthirtyyears,andwarrantsattentionfromsocial

scientists.Inthispaperwehighlighttheimportanceofunderstandingboth:1)

economicandpolicychangesinsendingcountries;and2)theincentivesthat

differentparentsmayhavewhenadoptingachild.Ouranalysisconsidersany

differencesintheadoptionprocessofchildrenabroadbyU.S.andforeign‐born

parents.Theestimatespresentedinhere,whileonlypreliminary,showthatthe

21

characteristicsofthechildrendifferaccordingtothenationalityoftheparent:while

U.S.bornparentstendtoadoptbabiesmostlyfromKorea,ChinaandRussia;

foreign‐bornparentstendtoadoptchildrenofallagesfromtheircountryoforigin.

Thissuggeststhattheincentivesandreasonstoadoptmayverywelldifferacross

parentsinwaysthatwedonotyetunderstand.

22

References

ChildWelfareLeagueofAmerica.1999.AdoptionFacts(accessed02.15.2009athttp://www.cwla.org/programs/adoption/adoptionfactsheet.htm

Bachrach,ChristineA.(1983)AdoptionasaMeanstofamilyFormation:Datafrom

theNationalSurveyoffamilyGrowth.TheJournalofMarriageandFamilyVol45(4)Nov1983pp.859‐865.

Brooks,Devon,SigridJamesandRichardP.Barth(2002)PreferredCharacteristics

ofChildrenintheNeedofAdoption:IsthereaDemandforAvailableFosterChildren?TheSocialServiceReviewVol76(4)Dec2002pp.575‐602.

Kossoudji,SherrrieA.2008.NewChildrentotheUnitedStates:ForeignBorn

AdoptionsintheNewImmigrantSurvey.Mimeo.Kreider,RoseM.2003.AdoptedChildrenandStepChildren:2000.Census2000

SpecialReports.Plug,Erik(2004)EstimatingtheEffectofMother’sSchoolingonChildren’s

SchoolingUsingaSampleofAdoptees.AmericanEconomicReviewVol94(1)March2004pp.358‐368.

Sacerdote,Bruce(2007)HowLargearetheEffectsfromChangesinfamily

Environment?AStudyofKoreanAmericanAdoptees.QuarterlyJournalofEconomicsVol121(1)Feb2007pp.119‐158

U.S.HomelandSecurity.VariousYears.TheHandbookofImmigrationStatistics.

23

Figure1:

24

Figure2:

25

Figure3:

26

Figure4:

27

28

29

30

31

32

Table1:FamilyCharacteristics(byParent’sBirthStatus) (1) (2) (3)

BothParentsU.S.Born

AtleastonParentForeignBorn

AdoptedfromSameCountry

AdoptedfromDifferentCountry

Father’sAgeAdoption 40.1 39.4 41.2

(0.018) (0.056) (0.062)Mother’sAgeAdoption 38.9 38.0 39.5

(0.016) (0.050) (0.055)ProportionCollege,Father 0.54 0.27 0.57

(0.001) (0.003) (0.004)ProportionCollege,Mother 0.58 0.25 0.52

(0.001) (0.003) (0.004)FamilyAnnualIncome $106,899 $68,231 $107,013 (236) (409) (808)AgeAdjustedIncomeDecile 7.1 5.6 6.9

(0.007) (0.018) (0.022)ProportionSingle,Mother 0.13 0.13 0.10

(0.001) (0.002) (0.002)ProportionSingle,Father 0.03 0.04 0.02

(0.000) (0.001) (0.001)ProportionwithBiologicalChildreninHH 0.15 0.20 0.19

(0.001) (0.002) (0.003)ProportionwithOtherChildreninHousehold 0.40 0.36 0.40

(0.001) (0.003) (0.004)N 6,770 1,263 711WeightedN 137,156 26,967 15,119Sample:Allchildrenunder18yearsofage,bornbefore2000inthe5%IPUMSCensus2000.Excludeschildrenwhomigratedatthesametimeoftheparent.

33

Table2.ChildCharacteristics(byParent’sBirthStatus) (1) (2) (3)

BothParentsU.S.Born

AtleastonParentForeignBorn

Same

CountryDifferentCountry

AgewhenAdopted 3.1 6.5 3.8

(0.009) (0.028) (0.028)ProportionFemale 0.61 0.55 0.60

(0.001) (0.003) (0.004)ProportionHispanic/Latino 0.19 0.48 0.21

(0.001) (0.003) (0.003)ProportionAsian 0.52 0.32 0.42

(0.001) (0.003) (0.004)ProportionBlack 0.03 0.09 0.06

(0.000) (0.002) (0.002)

N 6,770 1,263 711WeightedN 137,156 26,967 15,119Sample:Allchildrenunder18yearsofage,bornbefore2000inthe5%IPUMSCensus2000.Excludeschildrenwhomigratedatthesametimeoftheparent.

34

Table3.OLSRegressionDependentVariable:Child’sAgeWhenAdopted (1) (2) (3) ForeignBorn ForeignBorn U.S.Born Same

CountryDifferentCountry

GroupIntercept 7.816** 4.073** (0.973) (0.979)Mother,Age 0.192** 0.072** 0.130** (0.012) (0.018) (0.025)Mother,CollegeGrad ‐0.934** ‐1.855** ‐0.681** (0.141) (0.549) (0.351)SingleParent ‐0.048 0.501 0.299 (0.159) (0.465) (0.531)AgeAdjustedIncomeDecile 0.000 ‐0.006 ‐0.020** (0.003) (0.007) (0.006)OtherChildrenatAdoption ‐0.714** ‐0.909** ‐0.460** (0.101) (0.347) (0.205) Korea ‐0.095 (0.191) China ‐3.330** (0.170) Russia ‐1.443** (0.227) Mexico 0.582** (0.226) India ‐0.728* (0.388) Colombia 0.094 (0.201) Guatemala ‐1.586** (0.283) Romania ‐0.839** (0.224) Philippines 0.516* (0.313) R‐2 0.296 N 6,624 **Statisticallysignificantat5%*Statisticallysignificantat10%.AllregressionsincludeyearofimmigrationfixedeffectsSample:Allchildrenunder18yearsofage,bornbefore2000inthe5%IPUMSCensus2000.Excludeschildrenwhomigratedatthesametimeoftheparent.

35

Table4.LogitRegression,DependentVariable:ProbabilityChildisAdoptedatAgeThreeorOlder.Coefficientsrepresentmarginaleffects (1) (2) (3) Atleast1ParentForeignBorn USCitizen Same

CountryDifferentCountry

GroupIntercept 0.701** 0.525** (0.028) (0.049)MomAgeinUS 0.029** 0.005* 0.018** (0.002) (0.003) (0.004)MomCollegeGrad ‐0.119** ‐0.154** ‐0.092** (0.021) (0.054) (0.042)SingleParent ‐0.032 0.198** 0.084 (0.028) (0.073) (0.108)AgeAdjustedIncomeDecile ‐0.001* ‐0.001 ‐0.004** (0.000) (0.001) (0.001)OtherChildrenwhenAdoption ‐0.028* ‐0.086** ‐0.020 (0.015) (0.034) (0.036) Korea ‐0.093** (0.027) China ‐0.438** (0.017) Russia ‐0.121** (0.032) Mexico 0.066* (0.034) India ‐0.127** (0.053) Colombia ‐0.017 (0.034) Guatemala ‐0.241** (0.030) Romania ‐0.027 (0.047) Philippines 0.134** (0.048) PseudoR‐2 0.180 N 6,624 **Statisticallysignificantat5%*Statisticallysignificantat10%.Allregressionsincludeyearofimmigrationfixedeffects

36

Sample:Allchildrenunder18yearsofage,bornbefore2000inthe5%IPUMSCensus2000.Excludeschildrenwhomigratedatthesametimeoftheparent.