Intercountry Adoption in the ne= w Millennium:

the “quiet migration̶= 1; revisited

Paper presented at the European Population Conference

Hel= sinki, Finland 7-9 June 2001

REVISED & CORRECTED VERSION* - November 2002

Peter Selman =

Reader in Social Policy

Uni= versity of Newcastle upon Tyne

UNI= TED KINGDOM

* The paper was published in 20= 02 in Population Research & Policy Review, 21: 205-225,

The author can be contacted at;

School of Geog= raphy, Politics & Sociology, University of Newca= stle,

Claremont Bridge= level 5

Newcastle upon Tyne NE1 7RU,

ENGLAND<= /st1:country-region>. &= nbsp; Tel +44 191 222 7538; =

Fax +44 191 222 7497;&n= bsp;

e-mail p.f.selman@ncl.ac.uk

&n= bsp;

Introduction

Child adoption is not usually seen as a matter of concern for demographers, but rather an issue of primary interest to social workers, lawyers and psychologists and of secondary inte= rest to sociologists and anthropologists. The requirements for a demography of <= u>domestic adoption in Britain and the United States was considered in an earlier pape= r by the author (Selman 1997) presented at a meeting of the International Sociological Association in Beijing and has been discussed elsewhere in gre= ater detail in respect of adoption trends in England & Wales (Selman 1976 and 1987) and in a cohort analys= is of adoptees’ access to birth records (Selman 1999b)=

This lack of demographic anal= ysis is also true of intercountry adoption (ICA), although a recent article in t= he International Migration Review (Lo= velock 2000) looks at intercountry adoption as a migratory process and an earlier article in the same journal referred to international adoption as “the quiet migration” (Weir 1984).

Consideration of the movement= of children as a form of migration can also be seen in the long sad tale of “child migrants” from the UK (Bean & Melville, 1989), which continues to be a major issue both in th= e UK and “receiving states” such= as Australia and New Zealand. Most of the chil= dren were not placed in families and many were subsequently subjected to exploitation and abuse. But the placements were arranged by reputable child= care agencies and at the time were seen as in the best interests of the child. Commenting on this, the House of Commons Health Committee Report concluded = that “child migration was a bad and, in human terms, a costly mistake̶= 1; and urged “extreme caution” when considering applications for intercountry adoption.

Towards a demography of intercountry adoption<= /span>

In his 1984 article, Weil argued that many aspects of intercountry adoption we= re not well understood and concluded that :

“To answer questions such as those listed above requires far more data be

collected on a systematic worldwide basis. The total volume of foreign

adoptees, how this number has changed over time, precisely what countries

are linked in the flows of children….” (Weil 1984 pp 289-90)

The aim of this paper is to address some of these issues and give a clearer account of the growth of intercountry adoption over the past thirty years, using available national statistics from receiving states. I shall show that the scale of ICA is greater tha= n is usually acknowledged and could potentially grow in the first decades of the next century, making international controls even more important. I shall end by looking at some of = the social and demographic characteristics of sending and receiving countries.<= /span>

In doing so, I shall argue th= at a demography of intercountry adoption must see the movement of children not o= nly as an aspect of international migration - in which case it is of very minor significance - but also as related to issues of fertility, in t= hat a key motivation in receiving States is the demand for children by childless couples who have not been successful with infertility treatment and who have faced a diminishing availability of young children for domestic adoption (Hoksbergen in Selman, 2000). For this reason, it can be useful to relate intercountry adoptions to the number of births in both sending and receiving countries,

While = ICA was a phenomenon involving only a s= mall number of children from

relatively few countries, the possibility (or necessity)= of a demographic approach was

largely ignored. Today it is a phenomenon involving over 30,000 children a year

moving between over a hundred countries. Just as domestic adoption has been

shown to affect a large proportion of people in countries such as the USA (Hoksbergen, 2000), so most countries today have been affected by intercoun= try

adoption, whether as states of origin, receiving states = or as both (Selman 1997; van

Loon, 1990). This was the background to the establishmen= t of the Hague

Conference (Duncan, 2000), which was to involve over 80 countries during its

five years.

If we are to understand this phenomenon, it is important not only to be able to give an accurate picture= of the number of ICAs taking place and of changes in that

number over time, but also to identify those countries m= ost involved. This paper aims to estimate the number of intercountry adoptions world-wide, using data record= ed by 17 receiving states. Results will be compared with a similar exercise carried out ten years ago by Kane (1993), using data from 14 receiving countries.

Adoptions from these countries are believed to account f= or over 90 per cent of all international adoptions

The differing incidence of IC= A in States of origin will be estimated using data from 10 receiving States with reliable recording systems, which account for about 80 per cent of total ICAs, but sources of children= for adoption vary between countries so that the estimates for states of Origin = are distorted by the omission of key receiving states. The extent of this distortion is explored by checked the estimates against data available in t= hree sending countries (Korea, India and Brazil).

If we are to make comparisons between countries, whether receiving or sending, we also need to develop standardised measures of intercountry adoption as is routinely done for bir= ths, deaths and other demographic events (Selman 1999). In this paper I have chosen to fol= low Kane (1993) in standardising against the annual number of live births, but alternatives include calculating rates per 1,000 population (Pilotti 1990; Selman 1999) - or per 1,000 aged 5-9 for states of origin - and examp= les of these calculations will be included.

Problems of availability and accuracy of data on intercountry adoption=

&= nbsp; Weil (1984) noted that in the 1980= s “worldwide availability of d= ata on foreign adoptions is uneven in both quantity and quality” and that “ data from the United States, while in some respects limited, was the best obtained from any country” (Weil 1984 pp = 277-8) .

The availability and quality = of data on international adoption continues to vary greatly between receiving countrie= s. Kane (1993) approached government offices in 21 such countries, but was abl= e to obtain “relatively complete” data from 14 (see Table1): figures obtained for Canada were for Quebec only; and estimates for Germany were ba= sed on 4 northern lander; and statistics for Spain were only available from 199= 8. No statistics were obtainable from Austria<= /st1:country-region>, Israel, Ireland or the United Kingdom and three other (unspecified) countries faile= d to reply.

Such problems continue ten ye= ars later, despite the stress in the Hague Convention on the importance of gathering data systematically. In the Unite= d States, the Department of State has for many years published statistics on the numb= er of immigrant visas issued in each fiscal year to “orphans” comi= ng to the United States= . The U.S. Immigration and Naturalization Service (INS) has figures for the number of children actually entering the country on such visas. US statisti= cs are available on the internet, categorised by state of origin - so that it is possible to look at changes in the incidence of IC= A and also in the countries from which the children came.

= In Europe there is wide variation in the qu= ality and availability of statistics.

In Scandinavia and the Netherlands, domestic non-rel= ative adoptions are now very rare and only intercountry adoption statistics are published annually. In the Netherlands detailed statisti= cs on children adopted from overseas are provided by the Ministry of Justice. In Sweden there are annual figur= es from their National Board (NIA) on international adoptions by state of orig= in. In Norway the Governmental Office for Youth and Adoption provides information from 19= 79 on the annual number of child= ren mediated by the three licensed organisations. Detailed figures are also availabl= e for Denmark and Finland and for adoption agencies belonging to EurAdopt (see Sterky in Selman, 2000).

Annual figures providing information about cou= ntry of origin are available for France and Germany and in En= gland & Wales, the Department of Health provides the number of applications (home studies) received each year, but does not know how many actual adoptions take place. In the eight years fro= m 1993 to 2000 there were over 1,700 such applications (Brennan in Selman, 2000). = It is estimated that in addition over 100 children are being brought into the = UK each year without entry clearance. UNICEF (1999) provides total figures up to 1997 for seven countries including Canada, Italy and= Switzerland . I have also used incomplete= data for Spain (Catalonia only) and Belgium (adoptions through agencies which are members of EurAdopt) in order to match the 14 countries used by Kane. The overall data presented for these countri= es is more comprehensive than Kane’s for Germany and Canada, but less complete for Belgium= . In both exercises Spanish numbers are under-estimated. Three additional countries - UK, = Ireland and Iceland - have been incorpora= ted in the estimates of global numbers, the first two of which are also used (with eight receiving States) to estimate the number of ICAs in states of Origin<= /p>

The growth of intercountry adoption since 1970

The ear= ly history of ICA = has been described in many publications (e.g. Altstein & Simon, 1991; Selman, 19= 98; van Loon, 1990) . Altstein & Simon (1991) note that intercountry adoption “began primari= ly as a North American philanthropic response to the devastation of Europe in World War II that resulted in thousands of orphaned children”. European states were the main source of children to America until the late 1970s (Selman 1998) , when adoptions from Korea became increasingly important in num= bers, accounting for over half of all ICAs in America by the 1970s..=

Most estimates of global numb= ers in the early 1990s are for between 15,000 and 20,000 (e.g. Duncan 1993; Kane 1993; Lovelock 2000; = NAIC 1997) with an unstated assumption that there is little variation from year = to year. Yet the number of “orphans” arriving in USA alone has risen from 2080 in 1969 to 8,102 in 1989 and 16,396 in 1999, &nb= sp; If we accept estimates of 10,000 ICAs to Europe each year (Altstein & Simon 1991; Hoksbergen 1986) and note the growing numbers going to Ca= nada and Oceania, the total figures for the end of this century seem likely to be substantially higher (see also UNICEF,1999).

By far the best picture of intercountry adoption worldwide in the 1980s is that

provided by Kane (1993). Using data from 14 countries= , she calculates the

minimum number of ICAs between 1980 and 1989 at just over 162,000 - an

average of more than 16,000 a year. Noting that she was unable to get statistics for

ICA in the UK, Israel, Ireland and Austria, and that sta= tistics for Canada and Germany were incomplete, she concludes that there is a short= fall of 5-10 per cent and estimates the actual total for the decade as lying bet= ween 170,000 and 180,000 - an average of 17-18,000 per annum. <= /span>

More recently, UNICEF (1999) = has made an estimate of the number of adoptions to seven major receiving states (USA; Franc= e; Italy; Canada: Sw= eden; Switzerland and the Netherlands) in the period 1993-1997. This shows a sharp rise in numbers= over this period from 16,027 to 23,199, so that by 1997 the total for these seven countries is at least 5,000 higher than the world average estimated by Kane= for the 1980s. =

= Table 1: = Major Receiving Countries 1980 - 1998

&= nbsp;

Country

Mean annual adoptions

1980 -1989

1988

Mean annual

adoptions

1993 -1997

1998

United States

France

Italy

Canada

Spain

Germany

Sweden

Netherlands

Switzerland

Norway

Denmark

7,= 761

1,= 850

1,= 006

&nbs= p; 109¹

&nbs= p; 19²=

&nbs= p; 189²<= /o:p>

1,= 579

<= span style=3D'mso-spacerun:yes'> 1,= 153

&nbs= p; 616

&nbs= p; 464

582=

9,120

2,441

2,078

23= 2¹<= /p>

&nbs= p; 93²=

87= 5³<= /p>

1,074

57= 7

49= 2

56= 6

52= 3

10,070

3,216

2,047

1,934

&nbs= p; 784=

<= span style=3D'mso-spacerun:yes'> &nbs= p; 836

&nbs= p; 906

&nbs= p; 640

&nbs= p; 761

&nbs= p; 531

&nbs= p; 510

<= span style=3D'mso-spacerun:yes'> 15, 774

3,= 777

<= span style=3D'mso-spacerun:yes'> 2,= 263

2,= 222

1,= 497

^&nbs=
p;922

&nbs= p; 928

&nbs= p; 825

&nbs= p; 686

&nbs= p; 643

&nbs= p; 624

15,328

18,071

22,235

30,161

Belgium

Australia

Finland

544

356

662

516

183⁴

247

134

&nbs= p; 254⁴<= /o:p>

&nbs= p; 245

&nbs= p; 181

Sub-total

16,268

19,327

22,799=

30,841

New Zealand

Greece

Israel

Ireland

Luxembourg

Cyprus

Iceland

18= 0

23= 6

--=

61=

67=

--=

11=

38= 9

25= 8<= /p>

^&nbs=
p;236⁵

21= 4⁵<= /p>

14= 7

&nbs= p; 60

&nbs= p; 34

&nbs= p; 18<= /b>

TOTAL

23,354

32,197

&= nbsp;

Sources: Ka= ne (1993); Lehland (2000); Selman (2002) =

1. &n= bsp; For 1980-89, Canadian figures are for Quebec only (Kane 1993)

2. &n= bsp; Underestimate due to incomplete data (Kane 1993)

3. &n= bsp; Estimate based on 4 northern lander (Kane 1993)

4. &n= bsp; For 1993-8, “EurAdopt” adoptions only - total =3D 600+ (Lehland 2000)

5. &n= bsp; Figures for Greece and Israel are for 1994 and 1999 respectively

NB Total numbers for 1= 998 will be higher as Belgian figures are underestimated and no data were obtai= ned on some receiving states - e.g. Austria - (? 1,000 + extra). Lehland (2000) suggests c 33,500.

In Table 1, I have tried to build on these two exercises= by extending and updating the UNICEF calculation to include the other seven countries used by Kane plus New Zealand, the UK, Greece, Is= rael, Ir= eland, Luxembourg, Cyprus and Iceland. Data from Belgium and Spain are incomplete: the Belgian data are only for agencies which are members of Euradopt; the Spanish data are only for Cat= alonia.

My estimates for the total nu= mber of ICAs in the 1990s using the 14 countries covered in Kane’s analysis show a substantial increase in overseas adoption - from 19,327 in 198= 8 to 30,841 in 1998, reflecting mainly the shar= p rise in children going to the United States. Kane’s figures are underst= ated for Canada, Spain and = Germany but more comprehensive than mine for Belgium. Addition of the furt= her eight countries mentioned above brings the total to over 32,000. The global figure will be substantially higher, if we allow for an underestimate of numbers to Belgium and= Spain and the omission of other receiving = States which have signed, ratified or acceded to the Convention (e.g. Austria and Portugal). =

The incidence of ICA seems to be ri= sing in most receiving states, so that the totals for the years 1999-2002 are likely to be higher. This give= s a very different picture from Altstein’s prediction - made in the aftermath of the Romanian influx, which he saw as temporary, that …“..as a long-term world-wide phenomenon whereby nonwhite child= ren from poor nations are transferred to families in rich, white nations ICA appears to be declining” (Altstein & Simon&nbs= p; 1991) .

Which countries receive most children?<= /h2>
=

Table 1 shows that for the la= st 20 years, the United States has been the largest recipient of children for adoption, but that Canada, France, Italy, Germany and Spain also receive substantial numbers of children and many of the smaller European countries receive numbers which a= re relatively greater in proportion to their population.

Table 2 reminds us that the number = of intercountry adoptions can fall as well as rise and that for some European states - notably Sweden and the Netherlands<= /st1:country-region> - the current level is well below that fouund in the early 1980s.=

Table 2: Annual Number of International Adoptions: USA, Sweden,

Netherlands and Norway: selected years 1970 - 1998

Country

1970<= /span>

1975<= /span>

1980<= /span>

1985<= /span>

1990<= /span>

1995<= /span>

1998<= /span>

USA

Sweden

Netherlands

Norway

2,409

1,150

192

115

5,633

1,517

1,018

397

5,139

1,704=

1,594=

384

9,285

1,560

1,138

507

7,093

965

830

500

9,679

895

661

488

15,774

928

825

643

= ;

= ;

Standardisation of <= st1:City w:st=3D"on">ICA against population or birth levels = in receiving States

= ;

If we wish to compare the lev= els of intercountry adoption in either sending and receiving countries, it is essential to develop some form of standardisation as would be routine for a= ny other demographic event - births, deaths, marriages, divorce etc - but is rarely found in the adopt= ion literature. The simplest standardisation is to relate adoptions to the population size - a Crude (Intercountry) Adoption Rate. This has been used to make comparisons between receiving states (Sel= man 1989 and 1999; Pilotti 1990) = and shows Sweden as having= a much higher rate than the United States.= Such a rate could also be calculated for states of origin, but could= be very misleading in making comparisons between states with different age-structures.

In earlier analyses (Selman 1999, 2000), I chose 100,000 as the base for th= ese rates (rather than 1,000 population as in Crude Birth and Death Rates) beca= use of the low level of adoptions compared with births and deaths. Rates were calculated for five receiving states for the years 1987 to 1995 and ranged = from 11.9 per 100,000 for Norway to 1.9 for Finland. The United States, despite the large numbers of ICAs had a rate of only 3.3. By 1998 the Norwegian rate had ris= en to 14.6 and the American rate to 5.7, but both remained well short of the peak= of 22.7 found in Sweden= in 1980.

An alternative is to relate t= he adoptions to the number of births (Andersson 1986, Kane 1993 ). I have called this an adoption ratio (Selman 1998 ), defining this as the number of adoptions per 1,000 live births. Kane refers to a “rate of adoption” per 100 births. Adoptions are seen as in some sense the equivalent to acquiring a child through birth (Andersson 1986). Because of the similarity of demographic characteristics (e.g age-structure= and birth rates), the relativities are the similar whichever measure is used for receiving states. In 1998 the adoption ratio in Norway was 11.2, which indica= tes more than one intercountry adoption for every 100 live births. In Sweden in 1978 the ratio was 17.4 per 1,000 - nearly two adoptions for every 100 live births - or equivalen= t to a rise of 0.2 in the crude birth rate.

<= /o:p>

Table 3:&nb= sp; Intercountry adoptions per 1,000 live births; 1998 and 1989

Selected Receiving Countries ¹

Country

No of Adoptions

1998

No of Births

(1,000s)

1998

Adoptions per

1,000 births

1998

Adoptions ²

Per 1000 births

1989

Norway=

Sweden=

Denmark

Switzerland

Canada=

France=

Netherlands

Italy<= /st1:country-region>

USA

Germany

Finland

Australia

UK

       &nbs= p; 643

       &nbs= p; 928

       &nbs= p; 624

         686=

     2,222

      3,777

       &nbs= p; 825

      2,263

    15,774

    922

       &nbs= p; 181

       &nbs= p; 245

       &nbs= p; 258

       &nbs= p; 57

       &nbs= p; 86

       &nbs= p; 63

       &nbs= p; 80

       34= 4

       71= 3

       179=

       512

    3,788

749

       &nbs= p; 57

       24= 5

       68= 9

&nb= sp;

      11.2

      10.8

        9.9

        8.6

        6.5

        5.3

        4.6

        4.4

        4.2

        2.4

        1.2

        1.0

        0.4

&nb= sp;

11.0

9.4

8.5

6.2

3.0

2.7

3.7

2.0

3.8

1.6

2.0

1.4

N/A

= 1.&n= bsp;      Spain= and Belgium have been omitted as figures for 1998 were incomplete.

= 2.&n= bsp;      Kane’s figures per 100 multiplied= by 10.

Table 3 contrasts the figure= s for 1998 with those provided by Kane f= or 1989. As with the crude adoption rates, standardisation against births shows the level of intercountry adoption to be substantially higher in Norway and Sweden than in the USA. In most cases the level is higher = than in 1989 - substantially so in the case of S= witzerland, France and the United States.

Which countries send most children?

= ;

Weil (1984) = shows that in the early post-war years the main countries of origin were those defeate= d in the war - Greece, Italy, Germany and Japan - but that from the mid 1950s the main source of children to the United States became Korea, which accounted for the largest number in total over the period 1948-62 .&= nbsp; Between 1963 and 1975, Korea became even more dominant, accounting for nearly 15,000 out of a total of 34,568. The next six years (1976-1981) saw 19,283 children moving from

Korea to the United States out of a total = of 35,229. Ecuador, Colombia, Phillipines

and India followed with the numbers from Europe fa= lling to very low levels.

= &= nbsp; This set the patt= ern for the 1980s, where Kane (1993) identified Korea, Colombia and India as the major sending countries, conf= irming the picture given by Pilotti (1990) using data from the United States, Sweden and Norway.

= For a short period in the early 199= 0s Romania became the largest single source of children for international adoption (Defence for Children International, 1991). In the <= st1:country-region w:st=3D"on">United States alone the numbe= r of intercountry adoptions rose by nearly 2,000 between 1990 and 1991, the incr= ease being entirely due to the Romanian adoprions. The total number of adoptions from Romania in the months following the fall of Ceausescu remains uncertai= n, but the DCI report cited above lists a total of over 4,000 children going t= o 22 different countries in the seven months from August 1990 to February 1991 a= nd figures as high as 10,000 have been suggested for the period from March 199= 0 to June 1991, when Romania called a halt to adoptions (Selman 1998).

&= nbsp; Be= cause of the difficulties involved in obtaining comparative data from many states= of origin, I have followed Pilotti (1990) and Kane (1993) in using data gather= ed by receiving states to provid= e an estimate of the relative levels of intercountry adoption in states of origin in the = mid and late 1990s. Such figures = can be misleading where states of origin have particular links with receiving countries not included , but the exercise is useful in indicating trends ov= er time and comparative levels of involvement in ICA.&n= bsp; By 1995 China a= nd Russia had emerged as the main sources of children both for the USA and many other countries (Selman 1999),

Table 4: &= nbsp; Major sources of ICAs: 1980-89, 199= 5 and 1998

<= span style=3D'mso-spacerun:yes'> [Adoptions to selected Western Countries]

= ;

Country

Annual

adoptions

1980-89*

Country

No. of adoptions

1995**

Country

No of adoptions

1998**

&nb= sp;

Korea

India

Colombia

Brazil

Sri Lanka

&nb= sp;

Chile

Philippines

Guatemala

Peru

El Salvador

Mexico<= /o:p>

Haiti

Poland

Honduras

Thailand

6,123

1,532

1,484

753

682

524

517

224

221

218

160

153

148

110

China

Korea

Russia

Vietnam

Colombia

&nb= sp;

India

Brazil

Guatemala

Romania

Phillipi= nes

&nb= sp;

Paraguay

Poland

Ethiopia

Bulgaria

Thailand

&nb= sp;

Chile

Mexico

2,559

2,145

2,014

1,523

1,249

970

627

574

558

427

360

301

297

232

222

142

131

Russia

China

Vietnam

Korea

Colombia

&nb= sp;

Guatemala

India

Romania

Brazil

Ethiopia

&nb= sp;

Bulgaria

Thailand

Poland

Philippines

Cambodia

&nb= sp;

Haiti

Ukraine

Mexico

&nb= sp;

5,064

4,855

2,375

2,294

1,162

1,143

1,048

891

443

438

347

333

326

322

307

248

237

210

* &= nbsp; Kane (1993) - adoptions to 13 recei= ving countries [those listed in Table 1 excluding Finland]

**&= nbsp; Adoptions to 10 receiving countries [USA, Franc= e, Germany, Sweden, Ne= therlands, Norway, Denmark, A= ustralia, UK and Ireland] -= =3D c 75-80 per cent o= f ICAs

<= /o:p>

Table 4 give= s the number of adoptions in states of origin using data from 10 receiving States= for 1995 and 1998 and also Kane’s estimate for the 1980s, which are based= on data from 13 receiving countries. Although = Korea continues to be an important source of children, the annual number has fall= en sharply and was overtaken by the number from Vietnam in 1998. Kane’s= top 15 countries accounted for about 80 per cent of all adoptions to the 13 sen= ding countries in her analysis. In 1998 adoptions from the 18 sending countries listed accounted for 85 per cent of all adoptions to the = 10 countries in my analysis. Each additional set of data from receiving countries improves the picture and makes the list less dependent on USA figures. The major proble= m with the countries I have used is the omission of Italy, which would have substantially increased the numbers from Brazil, In= dia, Romania and Russia; and of Spain, which would have increased the numbers from India and Colombia (UNICEF 1999).

&= nbsp;

Table 5 = ; &n= bsp; ADOPTIONS from KOREA, BRAZIL & INDIA

1994 to 1998=

= ;

Country

Korea

Brazil

India

Year

Official Data ¹

Estimate

[Selman 2001]

Official Data ²

Estimate

[Selman 2001]

Official Data ³

Estimate

[Selman 2001]

1993

1994

1995

1996

1997

1998

1999

2,290

2,262

2,180

2,080

2,057

2,249

2,355

2,145=

2,294=

1,655

1,281

991

848

717

637

627

443

1,134

1,128

1,236

990

1,026

1,406

1,293

970

1,048=

= ;

1. Provisional data from Ministry of Health & Welfare

2. Data from Immigration section of Foreign Ministry

3. Data from CARA ( in Selman 2000)

&= nbsp;

Table = 5 shows that the 1998 estimates for Korea are close to those recorded by that country, as the 10 countries listed inc= lude all those receiving significant numbers of children. In contrast the estima= te for Brazil <= /u>is about a third lower than their official figures as a result of not includin= g Italy, which takes a large proportion of t= he children adopted from Brazil (40% in 1994). Likewise India is underestimated by 20-25% as I have not included adoptions to Spain and Italy.

A closer inspection of table 4 shows clearly how much change there has been in the sources of children in = the past decade, with four of Kane’s top fifteen countries - Sri Lanka, Peru, El Salvador and Honduras - no longer featuring in the lists f= or 1995 and 1998 - and numbers from Korea and Chile falling substantially. The a= nnual number of adoptions from China to the United States r= ose from 61 In 1991 to 4,206 in 1998; the number from Russia rose from 324 in 1992 = to 4,491 in 1998. By 1998 these two countries dominate the ICA field largely due US adoptions, but Russian adoptions also increased sharpl= y in France, Germany and Sweden; and Chinese adoptions in Sweden and the Netherlands<= /st1:country-region>. Between 1995 and 1998, numbers rose in most of the listed countries: only <= st1:country-region w:st=3D"on">Brazil, the Philippines, Chile and Paraguay experienced a fall in the number of adoptions, s= o that the increases seem to reflect an overall rise rather than changes in the countries from which children are coming.

There are a number of differe= nt reasons for a reduction in numbers of ICAs

from states of origin, which suggest that there are like= ly to be further changes in rank

ordering in the years ahead. Examples include:

· Crisis countries where the social/economic situation has transformed e.g. Greece and Germany which were major send= ing countries after WW2, but are now receiving children.

· Countries which have moved to domestic adoption - = e.g. Sri Lanka. Korea and India all have policies leadi= ng in this direction, which have already significantly reduced levels of intercou= ntry adoption.

· Suspension of adoption by either side - e.g. Paraguay (or Romania temporarily in the mid-nineties). In future year= s we may see falls in the number of children from Vietnam = and Guatemala<= /st1:place> due to their governments’ response to reports of trafficking.

<= /o:p>

Standardised= rates for states of origin

The number of children under the age of 5 f= orm a varying proportion of the total population of different states of origin: 1= 5-16 per cent in Paraguay a= nd Guatemala; 11-13 per cent in India, Vie= tnam and the Philippines; 7= -8 per cent in China and Korea; but only 5 per cent in Russia, Romania and other Eastern Eur= opean countries. For this reason a = crude inter-country adoption rate (per 100,000 total population is less suitable = for states of origin, where it is the number of young children which is most relevant in assessing the impact of ICA. Analysis of adoption rates per 100,000 population aged 0-4, using data from= ten receiving countries (Table 6) shows that in 1998 Bulgaria, Romania and Russia had the highest rates, closely foll= owed by Korea and Guatemala<= /st1:place>. The large numbers of children moving from C= hina and India are seen to be modest in relation to the under-5s population in those countries.

Table 6: Adoptions per 100,000 children aged 0-4; 1998 (and 1995):

States of origin sending at least 200 children to the te= n ¹ countries

Country

Annual number of=

adoptions

1998<= /span>

Population under= age 5: (millions)

1998<= /span>

Adoptions per 100,000 aged 0-5

1998<= /span>

Adoptions per 100,000 aged 0-5

1995<= /span>

Bulgaria²<= /p>

<= st1:country-region w:st=3D"on">Romania

<= st1:country-region w:st=3D"on">Russia

<= st1:country-region w:st=3D"on">South Korea

[Paraguay] ³

<= st1:country-region w:st=3D"on">Guatemala

<= st1:country-region w:st=3D"on">Vietnam

<= st1:country-region w:st=3D"on">Colombia

<= st1:country-region w:st=3D"on">China

<= st1:country-region w:st=3D"on">India

347

891

5,064

2,294

-----³

1,143

2,375

1,162

4,855

1,084

0.38<= /span>

1.06<= /span>

7.00<= /span>

3.43<= /span>

0.76<= /span>

1.79<= /span>

8.76<= /span>

4.79<= /span>

98.57

115.62

91.3<= /span>

84.0<= /span>

72.3<= /span>

66.8<= /span>

----^3<=
/o:p>

63.8<= /span>

27,2<= /span>

24.3<= /span>

4.9

0.9

55.0<= /span>

39.2<= /span>

25.6<= /span>

57.3<= /span>

[50.1]

30.3<= /span>

14.3<= /span>

28.2<= /span>

2.3

0.8

Source: Population data from UNICEF Country Statistics; http://www.unicef.org=

1. = See notes to Table 4

2.<= span style=3D'font:7.0pt "Times New Roman"'> = Latvia and Lithuania sent less than 200 children in 1998 but have similar overall rates to Bulgaria, resulting from their low numbers in the under-5 population.

3. = Paraguay had no recorded adoptions to the ten countries in 1998

<= span lang=3DEN-GB style=3D'mso-ansi-language:EN-GB'>

As with receiving states an alternative is to standardise against births (an adopti= on ratio), which accentuates the gap between high and low birth rate countries - e.g. South Korea and India. Table 7 shows that a ratio per 1,000 births gives a similar standardised order to the rates in Table 6= . This measure allows comparison with similar measure for 1989 calculated by&nbs= p; Kane (1993). In making comparisons it must be noted that my figures = are based on only 8 of the 13 countries used by Kane (plus the UK and Ireland) - which together for about 80 per cent of the adoptions from the countries listed so that equivale= nt ratios would be about 25% higher. Despite this the unadjusted ratios are hi= gher for a majority of countries featured.

&= nbsp;

Table 7: Adoption Ratio ( per 1,000 live bir= ths) 1998 and 1989

States of origin sending at least 200 children to the te= n ¹ countries

Country

Adoptions in 1998

Births in 1998

(1,000s)

Adoptions

per 1,000 births

1998*

Adoptions

per 1,000 births

1989**

Bulgaria

Romania=

Russia<= /st1:country-region>

South Korea

Guatemala

Vietnam=

Colombia

Haiti

Cambodia

Poland<= /st1:country-region>

Ukraine=

Thailand

China

Ethiopia<= /p>

Philippines

Brazil

Mexico

India

347

891

5,064=

2,294=

1,143=

2,375=

1,162=

248

307

326

237

333

4,855=

438

322

443

210

1,084=

202

1,420

682

393

1,681

988

253

364

418

&nbs= p; 482

1,000

20,134

2,652

2,064

3,340

2,335

24,671

4.9

4.4

3.6

3.4

2.9

1.4

1.2

0.94<= /span>

0.84<= /span>

0.78<= /span>

0.49<= /span>

0.33<= /span>

0.24<= /span>

0.17<= /span>

0.16<= /span>

0.13<= /span>

0.09<= /span>

0.04<= /span>

<0.1

5.4

0.8

2.5

1.1

0.8

0.3

<0.1

0.4

0.5

<0.1

* = Based on adoptions to 10 receiving countries (see notes to Table 4)

** = Kane’s figures are derived from 14 countries

= ;

NB:<= span style=3D'mso-tab-count:1'> In 1989 the highe= st ratios included Chile = (3.0), Sri Lanka (1.0) and El Salvador

(1.0), none of which reached the minimum of 200 ICAs in 1998 u= sed as a

criterion for inclusion in the table, although the average num= ber of adoptions

from 1980-1989 were 524, 862 and 218 respectively. By 1998 adoptions to

Paraguay (ratio of 2.0 in 1989) had virtually ceased.

Table 7 shows dramatically the changing pattern of intercountry adoption over the past decade from the point of view of states of origin. In 1989, the five countries with the highest level of ICA standardised against births, were (in descending order) Korea, Chi= le, Colombia, Paraguay and Haiti. Only one of these R= 11; Korea - still features in the top 5 countries. The level of ICA has risen in Bulgaria,= Russia, Ro= mania, Guatemala, Vietnam and China, but has fallen in Korea, Colombia, the Philippines and Brazil. Chile, Paraguay and Sri Lanka, which had high levels of ICA in 1989 -- no longer fe= ature in the top eighteen countries.

The table only includes those state of Origin sending at least 200 children in 1998 - inclusion of = all sending countries would find other states of the former soviet Union such a= s Latvia and Lithuania with adoption ratios similar to those recorded for Bulgaria and Russia.

The adoption ratio in sending countries tends to be lower than in receiving countries. This is partly a consequence of the larger number (and larger population) of sending countries. As we indicated above, adoptions to Scandinavia have been equivalent to 1-2 per cent of annual bir= ths, while those from Bulgaria and Russia are equivalent to less than 0.5 per cent of births in those countries. However, the latter are underestim= ates and in the early 1990s the adoption ratio for Romania would have been the equivalent of 4-5 per cent of annual births, although the age-range of the children moving in the peak ye= ar would make such a standardisation of limited value.

The standardised measures also offer insight into the potential scale of intercountry adoption. If China had the same level of ICA (measured by an adoption ratio) as Bulgaria, the number of child= ren moving would rise to nearly 100,000!, nearly three times the current estima= ted total for all states of Origin.

Demographic influences on intercountry adoption

The most commonly cited “causes” of ICA are the crises= of war, famine and disease

which make it impossible for poor countries to provide f= or all their children. ICA

continues to be largely a move of children from poor to rich countries (Selman

1998). A Malthusian interpretation would see these crise= s as demographic in origin!

It is, however, evident that the major sources have not = been the poorest or highest

birth rate countries that patterns persist long past the “crisis” and that demand for

children is as also a key factor. It is instructive to n= ote that states of origin are not all

high birth rate countries facing Malthusian population growth, but include countries

with total fertility rates below that of the major recei= ving states (see Tables 8 and 9).

That South Korea, China, Thailand and the former Communist states of Eastern Europe are sending children to the = United States and Sweden, when they have birth = levels below replacement level must give pause for thought.

Table 8: = = Economic & Demographic Indicato= rs : 1998

Selected Receiving States

Country

Adoption Ratio

1997/8

Per capita
GNP (US $)

1997

Infant Mortality

Rate, 1998

Total Fertility Rate

1998

= ;

Norway

Denmark<= /st1:place>

Sweden

Switzerland

France

USA

Netherlands

Italy

Finland<= /st1:place>

Australia

11.2

10.8

9.9<= /o:p>

9.2<= /o:p>

5.3

4.2

4.6

3.9

3.2

1.0

36,100

34,890

26,200

43,060

26,200

29,080

25,830

20,170

24,790

20,650

1.9

1.7

1.6

1.5

1.7

2.0

1.5

1.2

1.7

1.8

&nb= sp;

Source: S= tate of the World’s Children 2000

&nb= sp;

Table 9: = Economic & Demographic Indicators: 1998

Selected States of Origin

Country

Adoption Ratio (1998)

Per capita
GNP (US$)

1997

Total Fertility Rate 1998

Infant Mortality

Rate 1998

Bulgaria<= /p>

Romania<= /st1:place>

Russia

South Korea

Guatemala

Vietnam<= /st1:place>

Colombia=

Haiti

Cambodia=

Poland

Ukraine<= /st1:place>

Thailand=

China

Ethiopia=

Philippines

Brazil

Mexico

India

4.9

4.4

3.6

3.4

2.9

1.4

1.2

0.94<= /span>

0.84<= /span>

0.78<= /span>

0.49<= /span>

0.33<= /span>

0.24<= /span>

0.17<= /span>

0.16<= /span>

0.13<= /span>

0.09<= /span>

0.04<= /span>

1,170

1,410

2,680

10,550

1,580

&nbs= p; 310

2,180

&nbs= p; 380

&nbs= p; 300

3,590

&nbs= p; 750

2,740

&nbs= p; 860

&nbs= p; 110

1,200

4,790

3,700

&nbs= p; 370

1.2

1.3

1.7

4.9

2.6

2.8

4.3

4.6

1.5

1.4

1.7

1.8

6.3

3.6

2.3

2.7

3.1

104

110

&nb= sp;

Sources: Population and Economic data from UNICEF country statistics

The economic disparities in the per capita GNP are of co= urse vast; $20-36,000 for the receiving countries; less than $4,000 for all the sending countries other than Korea ($10,500) and Brazil= ($4,790). Similarly, the differences in infant mortality are substantial: 4= -7 for the receiving states; up to 110 per 1,000 for the states of origin. However, it must be noted that of the eighteen countries listed only seven = had a GNP less than $1,000 for a year ( and only five a GNP below $500 ) in a y= ear (1998) in which UNICEF’s State of the World’s Children gives the average GNP for the fifty least developed nations as only $256. The major sending countries are no= t the poorest in the world and Ethiopia is the only representative from Africa.<= o:p>

Demographic pressures and intercountry adoption

Three “sending” countries have dominated the story of intercountry adoption in the

1990s: Romania, China and Russia. Romanian adoptions are thought to have

accounted for at least&= nbsp; a third of all intercountry adoptions in 1990/1 (DCI, 1991),

but this was short-lived as the Romanian government reac= ted to international

criticism. But since 1995 it is China and Russia who have dominated American adoptions, accounting for more than half of the record number of 16,396 in 1999.

All three countries have experienced particular demographic pressures to which intercountry adoption has seemed to offer a relevant - if minor= and inappropriate - response.

Romania after the fall of Ceausescu

The flood of children from Romania in the early 1990s was triggered by media

images of desperately overcrowded institutions, but the crisis in those institutions

had built up over the previous 25 years of rule by dicta= tor Ceausescu whose pro-

natalist policies had banned legal abortion. Following the fall of Ceausescu, =

abortion was legalised and within a year the number of recorded abortions was three

times the number of live births (Hord et al 1991). By 1995, the total fertility rate = (1.5) was well below replacement level and it has fallen further since (see Table= 7).

Despite this, the rate of adoption from Romania = has been rising in recent years - by 1996 Romania was fourth as a source of children in USA and for Euradopt agencies. Renewed cha= rges of trafficking in children (Newsnight, 2nd March 2000) may reverse this tre= nd, as Romania was shown to lack the control over ICA expected of a country whi= ch has ratified the Hague Convention, an issue discussed at the Special Commis= sion of 28 November -1 December 2000.&nbs= p; Planned new legislation is hoped to resolve some of these problems.<= o:p>

China’s One Child Policy

’s “O= ne child policy” has created a crisis in the rejection of girl babies. By 1990

there was talk of hundreds of thousands of “missing girls” as mal= e sex ratios

reached over 110 (Johansson & Nygren 1991). This has been variously attributed to infanticide, selective abortion and non-registration of births (sometimes associated

with de facto adoption). Johansson estimated that as many a= s half of the missing

girls were adopted intracountry.

Inter-country adoption increa= sed from 1990 and especially after the Adoption Law was implemented in April 19= 92 and had built up to 4,206 in USA alone in 1998,

Yet even if the number of adoptions from China were to rise to 10,000 = per

annum this would be barely significant when set alongside the total numbe= r of

annual births [21,726,000 in 1995] or the population under 5 [5 million in 1995].

Johnson et al (1998) note a rise in infant abandonment (predominantly female) in

recent years, citing official (under)estimates of 100-20= 0,00 a year with 8-10,000

domestic adoptions. There have also been reports of poor families “selling”

unplanned babies to richer couples in China.

= China charges prospective adopters a flat rate fee of $3,000 (which is said to be= a contribution towards improving services for children) - a sum more th= an three times the per capita GNP in 1998 (see Table 7), which may make it difficult for the coun= try to reverse the rise in overseas adoptions.

The Russian Federation after the fall of Communism

The recent rise of adoptions from Russia is associated with one= of the most dramatic

demographic reversals in recent times. Between 1989 and 1994, life expectancy fell

from 73 to 65 and the number of male deaths rose from 762,000 to 1,226,000: - deaths to me= n in their forties trebled over the same period. Like Romania, Russia has a low birth rate, but the recent rise in mortality has led to a situati= on in which annual deaths exceed annual births by 50 per cent. The Russian population is declining and in twenty years time there could be a chronic labour shortage. So can Russia afford to send its children to the old enemy? In one sense, no - but = nor can it afford the costs of the growing number of children in institutions. Children adopted from Russia include many abandoned by poor mothers and many of these have foetal alcohol syndrome (McGuinness, 1999). A similar story is found in other parts of the former Soviet Union such as Latvia, Lithuania and the Ukraine. Current projections for the latter country are for population to halve in t= he next fifty years (Gerasymenko S & O, 2001).

What these three examples indicate is that a demographic history of adoption opens up the possibility= of linking intercountry adoption to demographic crises in states of origin, as well as to demographic trends (e.g. a rise in legal abortion) in receiving countries. But such crises may then establish a pattern that is hard to rev= erse even when the initial crisis is over, as has been argued in respect of our final example, Korea= .

50 years of intercountry adoption from South Korea

Korea continues to be a sending country with one of the highest rates of ICA. Since intercountry adoption began = in 1955, more than 120,000 children have been placed for adoption in other countries. Over 70 per cent of these went to the United States, where the numb= er of Korean “orphans” entering the country peaked at over 6,000 in the mid-1970s. . Initiall= y, many of the infants placed were of mixed race - the fathers being US military servicemen - but by = the 1990s there were very few mix= ed race children placed for adoption. =

Since the Olympic Games of 19= 88 there has been constant talk of reduction and eventual end to ICA. In 1989, the Ministry of Social Af= fairs proposed a schedule which would have reduced the number of intercountry adoptions to 1,700 by 1995

and raised the number of domestic adoptions to 3,500. However, by 1998 ICAs were still above 2,000 a year and domestic adoptions below 1500. A new 20 year plan was announced in 1997 to phase out ICA by the year 20= 20.

South Korea today is a prospe= rous country with an high level of education and a low birth rate, but there is a continuing problem over stigma of unmarried parenthood and in the absence o= f a comprehensive welfare system, it is impossible for a poor single mother to = keep her child. Sarri et al (1998) argue that ICA has discouraged Korea from developing an adequate child welfare programme. The example of South Korea reminds us that the factors influencing ICA may change over time and that there may also be a factor of inertia which m= akes it difficult to stop intercountry adoption.

Conclusion

= &= nbsp;

I have shown above that the number of intercountry adopt= ions is much higher than many recent estimates and have argued that it is now at its highest ever level in g= lobal terms - confounding predictions from the early 1990s that ICA was a phenomenon that had peaked. It seems probable that the first years of the n= ew millennium are likely to see even more children moving across national boundaries. Intercountry adop= tion remains - as it has always been - predominantly a movement of children from poorer to richer countries.

The level of adoption is determined= by the demand for children in rich western countries and the availability of children in those countries afflicted by poverty and other ills (Lovelock 2000). Several commentators (= e.g. Weil, 1984; Hoksbergen.in Sel= man, 2000) argue that the nature of intercountry adoption has changed over time = and that the humanitarian motivation of the early years has given way to a dema= nd from childless couples. The p= icture emerging in the United States - with numbers doubling in the last five years - suggests that there is a growing&n= bsp; demand for young light-skinned healthy babies, which has led to a tr= ade in children from and to countries, in some of which regulation of intercountry adoption falls far short of even the minimal standards sought by the Hague Conventio= n.

Whatever the trends in indivi= dual countries, the recent rise in total numbers of intercountry adoptions makes= the need for continuing research on the “epidemiological parameters”= ; of the movement of children (Kane, 1993) and on the Alternatives for children = and birth families in the States of origin even more crucial than it was in the early 1990s.

REFERENCES

&nbs= p;

Altstein H. & Simon J. (1991) Intercountry Adoption; a multinational perspective,

= New&nbs= p; York: Praeger.

Andersson G. (198= 6) "The adopting and adopted Swedes and their

&= nbsp; contemporary society" in R. Hoksbergen, Ado= ption in Worldwide

= Perspective= , Lisse: Swets & Zeitliger.

Apparao H. (1997) = “International Adoption of Children: the Indian Scene” International Journal of Behavioural Development

Bean P. & Melville J. (1989) Lost Children of the Empire. London; Unwin Hyman.

Defence for Children International (1991) Romania: the adoption of Romanian =

children by foreigners, Geneva: DCI/ISS. - also in DoH (1992)

Duncan W (1993) “The Hague Convention on the prote= ction of children and co-

= operation in respect of intercountry adoptio= n Adoption & Fostering

= 17-3 pp 9-13

Gerasymenko S & Gerasymenko O (2001) Poverty in Ukraine as the factor in fertility

reduction, paper presented at European Population Conference, Helsinki, Finland, 7-9 June 2001.

Hoksbergen R A C [ed] (1986) Adoption in Worldwide Perspective, Lisse:

&= nbsp; Swets & Zeitliger.=

Hord C et al (1991) “Reproductive Health in Romania: Reversing the Ceaucescu

Legacy= ” Studies in Family Planning 22 (4) pp 231-239

Johansson S & Nygren O (1991) “The missing girls of= China: A new

= demographic account” Population & Development Review 17-1, pp 35-51

Johnson K, Banghan H and Liyao W (1998) “Infant Abandonment and Adop= tion

in China” Population & Development Review Vol 24 no 3, pp 469-510.

Kane S. (1993) = “The Movement of Children for International Adoption: an epidemiological perspective”, The Social Scie= nce Journal, 30-4, pp 323-339

Lovelock K (2000) &= nbsp; “Intercountry Adoption as a Migratory Practice”&nb= sp; International

Migration Review, 34 (3) pp 907 - 949.

McGuinness, T. (1999)   Risk and Protective Facto= rs in Children Adopted from the

        &= nbsp;   Former Soviet Union, poster presentation at International Conference o= n

=             A= doption Research, Minneapolis, 10-14 August 1999.

Pilotti D (1990) &n= bsp;        Intercountry Adoption; Trends, issues = and policy implications

for the 1990s Inter American Children’s Institute

Ressler E. & Boothy N. eds, (1988) Unaccompanied Children, NY: OUP.

Ryvgold A, Dalen M. & Saetersdal B. (1999)     Mine - Yours - = Ours and Theirs,

    =         Oslo; University of Oslo<= /span>

Sarri R, Baik Y and Bombyk M. (1998)   “Goal Displacement and Dependency in

    =         South Korean - United States Intercountry Adoption” Children & Youth

        &= nbsp;   Services Review, Vol 20 - 1 pp 87-114

Selman P. (1976)     "Patterns of Adoption in England &a= mp; Wales since 1959" Social Work Today<= /i> 7-7 1976.

Selman P. (1988)      = ;           Family Planning, Reviews of UK Statist= ical Sources Vol XXV, published for the Royal Statistical Society and the Economic & Social Research Council by Chapman & Hall, 1988.

Selman P (1989) &nb= sp;    “Inter-country Adoption: what can B= ritain learn from the

experience of other European countries?” In G. Room Toward= s a

European Welfare State., Bristol: SAUS.

Selman P (1997)      = The Demography of Child Adoption: paper presented at ISA Workshop on Population, Beijing, 11 Oct 1997.

Selman P. (1998) &n= bsp;   “Intercountry Adoption in Europe after the H= ague

Convention” in R. Sykes & P. Alcock (eds) Developments in European

Social Policy: Convergence and Diversity, Bristol: Policy Press

Selman P (1999) <= span style=3D'mso-tab-count:1'>     “The Demogr= aphy of Intercountry Adoption”. In A. Ryvgold,

M. Dalen & B. Saetersdal Mine - Yours - Ours and Thei= rs, University of Oslo<= /st1:PlaceName>.

Selman P (1999b) In Search of Origins; twenty two years of access to Birth Records i= n

&n= bsp;           the UK. Poster presentation, International Conference on Adoption Research,

        &= nbsp;   Minneapolis, August 1999

Selman P (2000)    “The demographic history of intercountry adoption” pp 15-37 in

        &= nbsp;   P.Selman      (ed) Intercountry Adoption: Developments, Trends and

    =         Perspectives.        &= nbsp;   London: BAAF

Selman P. & Wells S. (1996)         = "Post adoption issues in Intercountry

        &= nbsp;   Adoption" pp 192-210 in E. McWilliams & R. Phillips (eds) After Adoption -

            Working with Adoptive Families. London: BAAF.

Selman P. & White J. (1994)        "The role of 'accredited bodies' in

        &= nbsp;   intercountry adoption" Adoption & Fostering 18-2 pp 7-13, [ Reprinted =

        &= nbsp;   in M. Hill and M. Shaw Signposts in Adoption, London: BAAF, 1998]<= /p>
UNICEF (1999)   Intercountry Adoption Information Portfolio

    =         http://www.unicef-icdc.org/informat= ion/portfolios/intercountry-adoption

Van Loon H (1990) Report on Intercountry Adoption. Hague: = The Hague

Conference.=

Weil R H (1984) &nb= sp;            =       “International Adoptions: The Quiet Migration”

International Migration Review, 18 (2) pp 276 - 293.