Migration Trends and Migration Policy in Europe Peter Stalker ABSTRACT This paper summarizes information on both stocks and flows of migrants in Europe, focusing specifically on arrivals from developing countries. It starts out by setting this into its historical context by showing how flows of labour migrants were followed by flows of family members, and later by asylum seekers and refugees. Then it looks more closely at recent migration data, though it finds these to be frequently incomplete and inconsistent. The most comparable cross-national data come from the OECD and Eurostat, which indicate that Germany had the largest flows of migrants in the 1990s followed by the United Kingdom. In addition to these arrivals there are probably between 2 and 3 million undocumented immigrants in Europe -accounting for 10 to 15 per cent of the total population of foreigners. The paper also traces the countries from where migrants are leaving. Sources vary considerably from one immigration country to another, reflecting a • number of factors, of which the most important are former colonial links, previous areas of labour recruitment, and ease of entry from neighbouring countries. In recent years, however, immigrants have been coming from a wider range of countries and particularly from lower-income countries. The paper also examines changes in immigration policy. National policies were fairly liberal during the 1950s and 1960s, before becoming more restrictive from the 1970s on. Recently, however, a number of governments have been revising their policies to take better account of employment and demographic needs. The paper also traces the emergence of a cross-national European response to immigration, as European Union (EU) countries have become more concerned about their common external frontier. Thus far European countries have done little to try to control migration through cooperation with sending countries. They could, for example, direct Official Development Assistance to those countries most likely to send immigrants, though few appear to have done so in a deliberate fashion. The paper concludes that in the future immigration to the EU is likely to increase, both as a result of the demand for labour and because of low birth rates in the EU. In the short and medium term many of these requirements are likely to be met by flows from Eastern Europe, particularly following the eastward expansion of the EU. But, the longer-term picture will probably involve greater immigration from developing countries. MIGRATION TRENDS AND MIGRATION POLICY IN EUROPE If it were possible, an aerial snapshot of migrant flows across Western Europe in the early years of the twenty-first century would offer a complex and confusing picture. For the purposes of international comparison, the simplest form of classification of these diverse flows is by four broad categories of entry. First, labour migration, which would include long- and short-term immigrants and seasonal workers. Second, family reunification, which usually consists of close relatives of those with long-term settlement rights. Third, undocumented workers or "illegal immigrants" who have either entered the country illegally or have entered on tourist visas and have overstayed, usually in order to work. Fourth, asylum seekers who, once granted asylum, are classified as refugees. To track these different flows and set the context for modern migration, a convenient starting point is the end of World War II. Since then, Europe has had four main phases of immigration. Late 1940s and early-1950s - mass refugee flows The end of World War II saw dramatic population shifts as around 15 million people transferred from one country to another, many of whom were forced to relocate as a result of boundary changes, particularly between Germany, Poland, and the former Czechoslovakia. By 1950, refugees made up 30 per cent of the population of West Germany (Borrie, 1970). From the mid-1950s these flows started to slow, though they continued at lower levels until the building of the Berlinwallinl961. Early-1950s to 1973 - recruitment of contract workers The reconstruction of Europe ushered in an economic boom. Between 1950 and 1973 the economies of the OECD countries grew on average by 5 per cent per year. This created a huge new demand for workers, and Germany, France, and the UK started to run short of labour. At first they were able to recruit many of those displaced during the war. Then they looked to other European countries that had been slower to industrialize, including Italy, Portugal, and Spain. But as these countries too became more prosperous recruiters had to look further afield. Some countries drew on their colonial ties. France turned to North Africa, and the UK to the Caribbean and the Indian subcontinent. Germany, without a colonial reservoir instead recruited short-term contract workers from countries adjacent to Western Europe, notably the former Yugoslavia and Turkey. Over this period net immigration for Western Europe reached around 10 million (compared with net outflows of 4 million for the period 1914 to 1949) (Stalker, 1994). 1974 to mid-1980s - the doors close Opposition to the arrival of large numbers of immigrants had already been growing in the late 1960s. In the 1960s this caused the UK, for example, to cut back the number of people who could come from the British Commonwealth. But it was the recession following the oil shock of 1973 that signalled a more general reversal across Europe and all governments effectively closed the doors to further labour immigration and expected guest workers to leave. These workers had, however, by now put down roots and preferred to stay. Even so, most governments shied away from the kind of punitive measures it would have taken to expel them and allowed family members of existing immigrants to join them. Before and during this period migrants had also started to choose from a wider range of destinations including Italy and other countries of southern Europe. The economic stimulus of joining the European Community also made Greece, Portugal, and Spain more attractive to immigrants. Mid-1980s to 2001 - asylum seekers, refugees, and illegal immigrants This was a period of political upheaval, particularly in Eastern Europe during and after the collapse of communism. Eastern Europeans, with more freedom to travel, started to join the thousands of people fleeing conflict elsewhere in the world and sought asylum in Western Europe. But others who formerly might have travelled as contract workers were also deflected to the "asylum door". This phenomenon had been evident as far back as 1980 when some 108,000 Turkish citizens applied for asylum in West Germany. From 1989-1998, more than 4 million people applied for asylum in Europe, 43 per cent of whom came from elsewhere in Europe, 35 per cent from Asia, and 19 per cent from Africa (Salt, 2000). As the pressure grew, however, Western European governments started to tighten up on asylum applications, and more people tried to enter illegally, either travelling on their own initiative or with the help of smugglers. Stalker Measuring migrant flows Each country has developed a system of migration measurement based on its own particular requirements. This can make it difficult to aggregate data across Europe or make valid cross-country comparisons. The most up-to-date information typically comes from those countries that maintain full population registers - requiring that both nationals and foreigners register with the local authorities. But even these registers are not exactly comparable since they can be based on a different duration for the minimum length of stay required to qualify as an immigrant rather than as a visitor—three months in Belgium and Italy, for example, but 12 months in Ireland. Moreover, some countries classify asylum seekers as immigrants while others do not. Data for Germany, for example, include some asylum seekers, but not all. Those countries that do not maintain population registers have to rely on other sources. With this in mind, Table 1 combines the latest data available from the OECD's continuous reporting system on migration, SOPEMI, and from Eurostat (Eurostat, 2000), showing the extent of migrant inflows during the 1990s. As Table 1 indicates, Germany occupies first place, partly because of the inclusion of some asylum seekers, and has a peak in 1992. The United Kingdom is next, though it follows a different pattern with a steady overall rise. Data on emigration are sparser and even less reliable than those on immigration; some countries, including France, do not collect such information at all. The latest available information on emigration is collected in Table 2. Again Germany accounts for the bulk of the flows, and comparison with Table 1 shows that in 1997 and 1998 emigration exceeded immigration. To complete the picture, since these data may not include flows of asylum seekers and refugees, is it also useful to present these as a separate category. In the earlier years, Germany was the main destination but by the end of the decade the United Kingdom had taken the lead (see Table 3). Combining outflows and inflows should give net migration, which will be positive if immigration exceeds emigration. But since emigration data can be missing or unreliable it may be better to arrive at net migration from another direction. One option is to monitor changes in overall population size, treating net migration as a residual. So the difference between the population at the beginning and end of th' year minus the difference between births and deaths can be taken as net migration. Eurostat uses this approach to estimate net migration rates for the 15 countries of the European Union (EU) (Eurostat, 2002). These are shown in Figure 1, which shows that net migration for this group of countries peaked in 1992, fell until 1997, and then started to rise again. 'viigraiiun irnnus uigiumsn yuu<.y TABLE 1 INFLOWS OF FOREIGN CITIZENS TO SELECTED COUNTRIES 1990-1999 (Ľiousands) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 AustrljT" _ - - - - - - 57 - - Belgium' Si 54 55 53 56 53 52 49 51 58 Denmarfc* 15 10 17 15 16 33 35 20 21 — Finland* 7 12 10 11 a 7 S 8 8 3 Franca" 102 110 117 99 92 77 76 102 138 104 Germany 642 921 1,208 9B7 77*1 7ĚS 70S 615 606 674 Greece"" 10 Ú 10 ta 15 18 20 17 35 - Hungary* 37 33 15 16 13 13 13 12 12 15 Ireland" - - - - 13 14 22 24 2\ 22 Italy— 57 71 53 51 53 68 143 - 111 268 Luxembourg" 9 1Ü 10 Ů 9 10 9 9 11 12 Netherlands' Et E* 33 aß ee 67 77 77 62 78 Norway" K 16 17 22 1B 17 17 22 27 32 Portugal" - - 14 10 6 5 4 3 7 11 Sweden' 53 44 40 55 75 36 29 33 36 35 Switzerland' 101 110 113 104 92 ee 74 73 76 86 United Kingdom" - — 204 190 194 206 216 237 2sa 277 Sources: 'SOPEMI, 200.1, based on national population regisier; "SOPEMI, 2001, source Other than population register; *'*Eurostat, 2000; ""lor 1990-199S. SOPEMI, 2001, based on national population register Tor 1998-1999, SOPEMI. 2001, source other tfian population register. Migrant stocks The other way of assessing the extent of immigration is to consider the total number of resident migrants, the "stocks". Data on stocks usually come from population registers of various kinds as well as censuses, though the UK arrives at this information indirectly from a regular sample survey of the labour force. Again there are differences in the way these data are collected. Most European countries make regular estimates of the citizenship of their populations. These, will indicate the number of "foreigners" but they will not include all immigrants since some will have naturalized and by definition have ceased to be foreigners. The only way to count the stock of immigrants is to estimate the number who are "foreign born". Some countries gather this information as part of census surveys though they may not do so very regularly. Table 4 collects some of the most recent data on the proportion of the population who are foreign citizens, along with the proportion who are foreign born. 164 TABLE 2 OUTFLOWS OF FOREIGN CITIZENS FROM SELECTED COUNTRIES, 1990-1999 (thousands) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Austria*" — - - — - - - 50 — - Belgium* 27 35 28 31 34 33 32 35 36 36 Denmark* 5 5 5 5 5 5 6 7 8 - Finland* 1 1 2 2 2 2 3 2 2 2 Germany* 466 498 615 710 622 561 559 637 639 556 Italy*** 7 6 7 - - - 8 8 - - Luxembourg* 6 6 6 5 5 5 6 6 7 7 Netherlands* 21 21 23 22 23 22 22 22 21 21 Norway* 10 8 8 11 10 9 10 10 12 13 Sweden* 16 15 13 15 16 15 15 15 14 14 Switzerland* 60 66 80 71 64 68 68 63 59 58 United Kingdom" 95 102 94 89 82 74 77 94 88 130 Sources: *SOPEMI, 2001, based on national population register; "SOPEMI, 2001, source other than population register; ***Eurostat, 2000. Although the data are not for matching years, they do confirm the expected result that there are more foreign born than foreign citizens. This difference will be greater in countries where it is easier to gain citizenship - as in France where the proportion of the population who are foreign born is 11 per cent but the proportion who are foreign citizens is only around 6 per cent. For most countries there were no significant changes between 1990 and 1999. The largest increases seem to have been in Austria and Denmark. Elsewhere the proportion of people who are foreigners has been static or falling, though this could also reflect a combination of high levels of immigration combined with high levels of jnaturalization. The data in the middle two columns show the citizenship of the workforce. These roughly parallel the patterns in the first two columns. Illegal immigrants In addition to these immigrants there are also millions of other undocumented I workers - whose numbers are thought to have increased substantially during the 1980s and 1990s. Since illegal immigrants tend to avoid being registered, any estimates of the total number in Europe are necessarily guesstimates. One of the best indications comes from regularizations when a country declares an amnesty for certain categories of illegal immigrants who then have an incentive to come forward. Italy, for example, has had a series of regularization programmes. The latest of these, in 1998, attracted 3 50,000 applicants, which combined with data on legal immigrants, indicated a total foreign population of 1.6 million, of whom around 20 per cent were there illegally (SOPEMI, 2000). Spain's regularization programme in 2000 attracted 245,000 applicants which would indicate a 1 foreign population of 1 million, of whom around 25 per cent were there illej (SOPEMI, 2001). Italy and Spain probably have higher proportions of ill workers than other countries because they form part of Europe's sout border, are easier to enter, and are seen as transit countries for people hea further north. For Europe as a whole, the stock of illegal immigrants has 1 assumed to be between 2 and 3 million-which would constitute between 10 15 per cent of the total population of foreigners (IOM, 2000). If stocks of illegal immigrants are difficult to count it is even more difficu estimate the rate they are flowing into Europe. One of the most commonly qu flow estimates was made in 1994 by Jonas Widgren of the Vienna-b; International Centre for Migration Policy and Development. He started fron fact that, in 1993, 60,000 undocumented immigrants were apprehende European borders. After talking to many border control authorities he estim that perhaps four to six times as many people were not caught. This would rr that some 250,000 to 350,000 were getting through each year (Widgren, 19 TABLE 3 INFLOWS OF ASYLUM SEEKERS INTO SELECTED COUNTRIES, 1991-2000 (thousands) -------- 1991 1992 1993 1994 1995 1936 19&7 1395 1999 at Austria 27.3 T 6.2 4.7 5.1 5.9 7.0 6.7 13:6 20.1 B At 1( Belgium 16.4 17.6 26.5 14.7 11.7 12,4 11J 22.1 35.8 Daniu ark ■1.6 13.9 143 6.7 5.1 5.9 5.1 5.7 6.5 Finland 2.1 3.6 2.0 D.B ae 9.7 1.9 1.3 3.1 France 47.4 2S.9 Í7.6 36.0 20.4 17.4 21.4 22.4 39.9 3! Germany 255.1 43B.2 322.5 127.2 127.9 116.4 104.4 S3.6 95.1 7t Greeca s.r 2.0 0.8 1.3 1.4 1:6 4.4 2L6 T,5 : Ireland * ■ at 0.4 0.4 m 3.9 4-6 7.7 1C ftaly 26.5 6.0 1,6 1.6 1.7 0.7 1.9 11.1 33.4 t 4Ě Luxembourg 0.Z 0.1 0,2 0.2 a.2 0.3 0.4 1.5 2.9 Netherlands 21.6 20.3 35:4 52.6 29.3 Z2.B 34-4 45.2 «2.7 Norway 4.6 5.2 12.3 3.4 1.5 i.e 23 8.5 10.2 tc Portugal 0.2 DL5 T.7 0.5 0.3 0.2 0.3 0.3 0.3 c SľVi-i: S.I n.7 12.6 12.0 5.7 4.7 5:0 6-3 B.4 7 Sweden £7.4 64.0 37.e 16-6 9.9 6.6 9-6. 1^5 11.2 1É SwrtjBflanri United 41.6 16.0 £4.7 16.1 17.0 t6.0 24.0 it A 45.1 17 Kingdom 73.4 32. a 2B.Q 422 55.0 37.0 41.5 SSO 91.2 97 Tola! 559.2 ( iftg.5 £53.3 329.7 393-3. ' 254.9 276.1 356.4 447.1 418 Sources: SOPEMI, 2001 í I I FIGURE 1 NET MIGRATION RATE PER THOUSAND INHABITANTS EUROPEAN UNION, 1990-2000 4.5 4.0 3.5 3.0 H 2.5 2.0 1.5 1.0 0.5 0.0 4-1 3.7 2.8 2.9 2.9 2.1 2.2 2.0 1.9 1.6 1.8 1.3 U. L 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Source: Eurostat, 2002. Sources of immigrants The data indicate the destinations of migrants to Europe in recent years. But, from where have they been coming? The national composition of the immigrant population varies considerably from one destination country to another and reflects a number of factors, the most important of which are former colonial links, former areas of labour recruitment, and ease of entry from neighbouring countries. The simplest snapshot of source countries comes from data on the foreign bom. For this group of countries, 47 per cent of the foreign bom came from other European countries, 27 per cent from Africa, 10 per cent from the Americas, 14 per cent from Asia, and 2 per cent from elsewhere. There are significant differences, however, between the proportions for individual countries. Switzerland, Ireland, and Luxembourg, which have the highest proportions of EU nationals, did not have colonies, so they are a less obvious choice for immigrants from Asia or Africa. At the other end of the scale are former colonizing countries: France, the Netherlands, the UK, and Portugal. In France, for example, 32 per cent of the foreign bom came from Algeria and Morocco; in the Netherlands, 26 per cent came from Indonesia and Suriname; in the UK, more than 20 per cent came from the Indian subcontinent; and in Portugal, 49 per cent came from Angola and Mozambique (Salt, 2000). Migration Irenas ana migration policy in Europe 01 TABLE 4 STOCKS OF FOREIGN CITIZENS, FOREIGN WORKERS, AND FOREIGN BORN, 1990-2000 Foreign citizens as a. Foreign workers as a % kA total population1 % ol lotal workforce 1390 1999 199Q4 1999' Foreign born as a % of tolal population Vear % Austria Belgium Denmark Finland France Germftny Ireland Italy Luxembourg Netherlands Norway Portugal Spain Sweden SwlCůíland United Kingdom 5.9 9,2 7.4 9.5 9.1 8.3 7-1 B.7 3.1 4,9 2,4 2.5 0.5 1.7 - 1.2 6.3 5.6 e.2 e.i 6.4 B.9 7.1 S.7 2.3 3,1 2.6 3.4 1.4 2.2 1.3* 3,6 29,4 3E 45.2 57.3 4,6 4.1 3.1 3,4 3,4 4 2.3 2,9 1.1 1.9 1.0 1.8 0.7 2 D.6 1.0 5.6 e.S 5.4 4J 16.3 19 2 13.9 13.1 3.2 3.8 3.3 3,9 1999" 6.3. 2000- 2,5 19901 11,0 2000* 6.3 1991* 30.2 Enno2 fj.a 19991 6.5. 19911 4,6 1991J 2-2 2000" 11-1 1990? 21-3 I9997 75 Note: -For 1591. Sources: 1. SOPEMt, 2001; 2. Eurostai New Cronos dal abase, 2002; 3. Sali et 31 2000; 4. SOREM I. 2000. Spain is something of an exception in that the largest single source of foreign-bom residents came from France, not former colonies. This is probably because in the early 1990s immigration to Spain was still relatively low and the foreign bom were only around 3 per cent of the population. Belgium too is an exception; although Belgium had colonies in Africa, and during the 1960s and 1970s did require workers for its iron and steel industry, it largely recruited from other European countries, notably Italy (Stalker, 1994). Another way of looking at the same data is from the perspective of the sending areas. Of emigrants from Africa, for* example, 66 per cent went to France, and from Asia 55 percent went to the UK. For more recent trends a better indication of source countries comes from data on flows. The overall trend is shown in Figure 2 for eight of the countries of the EU (SOPEMI, 2001). This shows a steep rise in immigration, peaking in 1992-1993, followingthe collapse of communism and the breakup of the former Soviet Union, which provoked a sharp increase in migration to Germany, particularly of ethnic Germans. As these crises abated somewhat, however, and European Stalker countries became more restrictive, overall immigrant numbers started to fall, particularly for Germany. In the last few years, however, the total has been rising again. Table 1 also confirms that for a number of countries, including the Netherlands, Finland, and Sweden, immigration has remained fairly stable. Although the inflows appeared to fall in 1999, this is thought to be the result of a "technical adjustment". But other European countries have seen significant increases from 1998-1999, including Germany, Italy, and the UK. In Portugal immigration has also been rising, though from a lower base. FIGURE 2 MIGRATION TO SELECTED EUROPEAN COUNTRIES, 1984-1999 2,000 t 1984 1986 1988 1990 1992 1994 1996 1998 2000 Note: Inflows of foreign citizens into Belgium, Denmark, Germany, France, Luxembourg, the Netherlands, Sweden, and the United Kingdom. Source: SOPEMI.2001. Migrants to Europe now seem to be entering via most of the immigrant categories, though the largest numbers are family members, followed by labour migrants and asylum applicants. Family reunification - This is often the largest category of legal arrivals. In Denmark, for example, around 66 per cent of those given residence permits in 1999 for more than 1 year were family members, while 16 per cent were workers, and 8 per cent were refugees. In Sweden the proportion of inflows of non-Nordic or European Economic Area (EEA)' citizens arriving for family reunification was nearer 80 per cent. In Portugal 47 per cent of official arrivals were family members, a similar proportion arrived on work visas, and the rest were refugees. In the UK, however, the proportion of family members is lower - ivilgrattun imnus unu mi£iunu»//u>iu> >.. ^».. ^h ■ around 43 per cent of non-EEA citizens (SOPEM1,2001). The nationality of the newly arriving family members naturally follows the pattern of previous immigration. In Germany and Switzerland, therefore, most of the joining family members come from Turkey and the former Yugoslavia. In France and to a lesser degree in Belgium and the Netherlands, they come from N orth Africa, and in the UK they come most from India, Pakistan, and Bangladesh. Labour migration - After several years of economic expansion, Europe has seen a notable increase in labour migration. Some countries have actively been recruiting again at the higher end of the jobs market, yet they have also started to take on more unskilled workers, usually on a short-term or seasonal basis, particularly in agriculture, construction, and manufacturing, and also in services such as hotels and catering. For Europe as a whole, the majority of non-EU short-term workers come either from Eastern Europe or from Africa (IOM, 2000). Refugees and asylum seekers - This is the most volatile category, ebbing and flowing, according to political and economic conditions. The largest flows in recent years were of refugees escaping the Kosovo crisis. Hundreds of thousands of Kosovans fled to the West in 1999, though by mid-2000 most of these had returned home. Table 3 shows the destination countries for asylum seekers in Europe. In 2000 the largest number of requests in Europe were going to the UK, followed by Germany, the Netherlands, and Belgium. Where are asylum seekers coming from? Table 5 shows the main sources of close to 1.8 million asylum applications in Europe during the years 1994 to 1999 (UNHCR, 2001). Despite the violence in Europe in 1999 the largest number of applications over this period came from Asia. A similar pattern is also evident in more recent data. For October and November 2001 the top four source countries of asylum in Europe were Iraq, Afghanistan, the formerYugoslavia, and Turkey. Diversity of immigrants Another trend in recent years is that migration flows have become more diverse. Although the main destination countries continue to receive the bulk of their immigrants from traditional sources, they are also seeing people arrive from a broader array of countries. To some extent this is the result of political instability in many source countries, combined with falling costs of travel. One measure of this is to consider what proportion of the foreign population is accounted for by the top five immigrant groups. For most countries this proportion has tended to fall over the past decade. This is seen in Figure 3, which ranks countries according to the diversity of immigrant inflows in 1999. Of this group, the most diverse is the Netherlands where only 27 per cent of the 78,400 arrivals in 1999 came from the top five countries - in this case the UK, Germany, Morocco, Turkey, and the US. At the other end of the scale is Luxembourg where 55 per cent came from the top five countries: France, Portugal, Belgium, Germany, and the US. Ireland ranks in the middle of this figure, but in this case the proportion refers to only two countries: the UK and the US. TABLE 5 SOURCE COUNTRIES OF ASYLUM SEEKERS TO EUROPE, 1995-1999 Origin 1995 1996 1997 1398 1999 Total Yugoslavia, F P 51759 38,451 46,401 96.270 120,614 357,435 Iraq 18,198 26r23B 40,434 40.821 35,129 180,370 Turtey 41,335 38,416 33,106 21,762 19.723 154,392 Afghanistan 11.669 12,513 16.343 13.633 24,220 83,376 Sri Lanka 12,765 13,004 14.118 12,345 12,735 65.097 Iran 10.994 11,644 10,148 10.315 16,157 S9.258 Samalla 12,290 8.001 8,963 12,129 14,272 55,745 Romania 14,537 9.757 10,630 8,736 0,657 52,367 Bosnia and 17,231 E.4S4 8,226 10,207 6,679 48,327 HEraegov^na Pakistan 9,825 7,838 a,£70 6,596 e.163 40,692 Democratic Republic 7,761 7,847 a,7sa 6.959 7.249 38,542 ot Congo AkjBria a.67a 5.213 6,650 8.228 3.027 37.011 India 9,131 7,393 5.833 4.792 e,60fi 33.755 Armenia 5,746 8,957 6,036 5.345 3,645 32,779 Russian Federation 4,740 4,900 5.4ao 5,347 11,435 32.462 China 3.924 4.269 8,902 6,020 10,760 31,895 Nigeria 8£Q1 6.354 5.281 5,659 4,693 30,533 Albania 1,263 1.48S 7,S80 6,651 4,058 21,334 Stateless 3,120 3,550 3.650 3,661 4,227 18.208 Georgia 3,150 3,102 4,410 4.117 3,422 18,201 Alŕlca 71.854 55,631 58,950 64,040 72,862 323.337 Asia 143,835 146,103 165,216 150.294 171,77t 770,224 Europe 107.347 76,361 97,784 146.865 175.239 603,646 Latin Ameiiciui and 3,131 4,013 4.320 3,874 4,975 20,81 B Caribbean Other/unknrjwn 7.679 3,701 6.013 5,664 38,940 67,197 Total 333.346 292,81.3 332,783 370.937 463,337 1,794,222 Source: UNHCR, 2001. For the majority of countries the proportion represented by the top five countries has tended to fall in recent years. The most striking difference in this figure is for thi Netherlands, down from 49 per cent to 27 per cent, much of which is accounted for by declining numbers coming from Morocco and especially from Turkey, which was the leading source of immigrants in 1990 but had dropped to fourth place by 1999. Portugal too has seen a decline in immigration from Angola and Guinea-Bissau. Yet, some countries seem to have seen a greater con- centration in the immigrant population. In Norway, for example, the 1999 figures were affected by a large inflow of refugees from Kosovo, though there has also been a notable increase in the number of people moving in from Sweden. Apart from an increase in diversity, there has also been a tendency for the foreign population in Europe to come from lower-income countries. In the mid-1980s the majority of the foreign populations from EU and EFTA countries came from other high-income countries, and generally from other European states. The main exceptions were Germany and the Netherlands, which already had high numbers from Turkey, and Portugal which had many immigrants from its former colonies in Africa. But during the mid-1990s most countries that previously received a majority of their immigrants from high-income countries increased the proportion coming from lower-income countries. Even these are usually from the lower-middle income countries (per capita income in 1995 $766 to $3,035), rather than the very poorest countries in sub-Saharan Africa or elsewhere. Italy's increase, for example, included many more people from Morocco, Tunisia, and the Philippines (Salt, 2000). IMMIGRATIONPOLICY Patterns of immigration are also shaped by government policy which attempts to control immigration flows in the national interest. At its simplest this can be seen as an attempt to balance two conflicting objectives. On the one hand governments welcome immigrants as a valuable labour force, either as workers whose skills are in short supply, or as unskilled workers who are prepared to do some of the jobs that native workers shun. On the other hand they also try to dissuade immigrants if they believe they will bring social and political problems and they usually restrict immigration on the grounds of preserving "national identity" or maintaining social stability. Similarly there are two main conflicting factors when it comes to accepting refugees. The main reasons for accepting are social and political - responding to humanitarian impulse to create a safe haven for those who have a "well-founded fear of persecution". The primary reason for trying to limit the flows of refugees is usually economic, since refugees can be seen as a drain on publicly funded welfare services, though there are also concerns about social stability. In reality, of course the situation is far more complex, and these and many other factors interact and mutate. In the receiving countries, governments have to respond to pressures from many different interest groups, some in favour of liberal immigration policies and others demanding stricter controls. Thus employers can be expected to be in favour of immigration which gives them a larger pool of potential employees. Workers' groups, on the other hand, may object to competition from immigrants who are prepared to accept lower wages. Nevertheless during times of economic expansion even workers' epresentatives tend not to oppose immigration very strongly. Germany's guest vorker programme, for example, was worked out in close cooperation with trade inions(Hollifield,1992). FIGURE 3 PROPORTION OF INFLOWS OF FOREIGNERS FROM TOP FIVE COUNTRIES, EARLY AND LATE 1990s Netherlands i 1 ■ Early 1990s ■ Late 1990s Denmark Sweden ^i Germany i France J _ Belgium Switzerland í ii^^^^ Ireland i _____„______, Portugal ' UK —1 Norway ^i Finland : Luxembourg ---------1-------------- -------1— —I 0% 20% 40% 60% Proportion from top five countries 80% Notes: Data are for inflows for 1990 and 1999, except Denmark (1992 and 1998), Finland arid the UK (1992 and 1999), and Belgium (1991 and 1999). For Ireland they are 1994 and 1999 and the proportion is for the top two countries. Source: Data from SOPEMI, 2001. A further factor likely to favour more positive immigration policies is demographic change. In the EU, the total fertility rate, the average number of children a women bears in her lifetime, is now down to 1.4, far below the figure of 2.1 required to maintain a stable population. As a result there have been suggestions in recent years that European countries should encourage immigration to offset the ageing of their populations, though the scale of immigration required to achieve this is dramatic. To maintain a stable population, the UN estimates that from 1995-2000 the EU would need to boost its annual inflows by a factor of five, to around 1.6 million. And if it wanted to maintain a constant "support ratio", the number of working people for each person older than 65, the EU would need to achieve net annual migration of 13 million (UN Population Division, 2000). There are similar fluctuations in attitudes to refugees. In principle the number of asylum claims that are accepted should be determined only by the number of people who have a well-founded fear of persecution. In practice, however, the proportion admitted also depends on the overall number of claimants since the stringency with which claims are tested tends to rise with the prospect of more arrivals. National policies on immigration Ultimately immigration in Europe is controlled by national governments. These policies can be quite diverse but during the 1950s and 1960s they were fairly liberal. The countries of northern Europe that had colonial histories needed more workers and were happy to accept them from their former colonies. In the 1950s the United Kingdom, for example, allowed people from the former British Empire to come to the "mother country" and accorded them the same rights as any other citizen, though it curtailed these settlement rights from 1962 on. France, concerned about falling birth rates, also gave citizenship rights to people not just from parts of the Caribbean that were administratively part of France, but also to arrivals from some former colonies in North Africa. The Netherlands and Belgium also had colonial links but actually recruited more guest workers from southern Europe, Morocco, and Turkey. West Germany also allowed free immigration to some extent, notably for "ethnic Germans" either from East Germany or from other countries in Eastern Europe or the Soviet Union. Then during the 1960s, West Germany started actively to promote the temporary immigration of young male workers through inter-governmental agreements with other West European countries as well as with Turkey, Morocco, Tunisia, and the former Yugoslavia. Switzerland also recruited guest workers over this period but always on a strictly controlled basis. The position of the Nordic countries was somewhat different. From 1954, Sweden, Denmark, Finland, and Norway had established a common labour market, and from 1957 a common passport union, allowing people to work freely in each other's countries. This allowed Sweden, for example, to draw in workers from Finland, though later it also established a controlled system for foreign labour from other countries, including the former Yugoslavia, Greece, and Turkey. Here the first main attempt at control was from 1965 when workers had to obtain a work permit before arrival. Denmark also used guest workers during this period. Norway, during the 1950s and 1960s, had relatively little immigration. 74 Stalker Immigration policies that had been diverse suddenly converged after the oil crisis of 1973-1974. Most countries passed legislation to restrict further primary immigration. And countries with guest workers also encouraged them to return home. But such policies had little success. Many guest workers in Germany and elsewhere had already put down roots and started families. Although Switzerland did deport some workers, other countries were reluctant to take measures that could be seen to infringe human rights. A similar concern for human rights also ensured that immigrants already in place would be allowed to bring in close family members. The effect of the controls was therefore to shift inflows from labour immigration to family reunification. The European dimension This period also saw the slow emergence of a cross-national European response to immigration. Initially this was limited to agreements between members of the EU on travel and labour issues within European countries. But from the mid-1980s the countries of the EU became more concerned about their common external frontier and struggled to develop a common policy on non-EU immigrants. Some of the stages in this process are listed in Table 6. One of the most significant moves was the adoption of the Schengen accords, originally signed in 1985. In 1990 these were formalized into the Schengen Convention which moved the EU closer to a borderless union and to common policies on immigration and asylum. This involves removing border controls between EU countries while hardening external frontiers, creating what is popularly referred to as "fortress Europe". This has not yet been implemented completely. Denmark, for example, will decide on a case-by-case basis whether to participate. And the UK and Ireland are not parties but can, with the approval of the EU Council, apply the Schengen Convention in whole or in part and participate in its further development. On the other hand two non-EU countries, Iceland and Norway, are now part of the agreement. Schengen also offers some freedom for non-EU nationals who can move between signatory countries if they have the appropriate visa or residence permit. In the case of asylum seekers the most significant effort at establishing common frontiers has been the 1997 Dublin Convention which requires asylum seekers to apply in the first EU country in which they land. In practice this has proved very difficult to implement and relatively few people have yet been transferred (DRC, 2001). One of the most dramatic indications of its failure to work as planned is the gathering of asylum seekers at the French end of the Channel Tunnel hoping to gain asylum in the UK. TABLE S MIGRATION POLICY IN THE EUROPEAN UNION Vear" Event Outcome 195S Treaty of Rome Sel up the European Economic Community and established thai a rati zen o< one member country c cold travel to another country lo work or seek wofk- 1976 Trgvi Group A mooting of ministers to promote cooperation on issues of lew end order (TrovT now stands For Terrorisms, Radicalisms. Extremisme, el Violence Internationale) After 195S triia becomes an ad hoc group of ministers responsible lor considering immigration questions, particularly illegal Immigration, among elfter things. 19B5 Schengen Accords Art agreement to rámovo ad border controls while attempting to strengthen the common externa! frontier. Originally signed by six countries in 1985, tfte current signatories are Auslria, Belgium, Denmark, Finland, France, Germany, Greece. Iceland, Italy, Luxembourg, lite Netherlands, Norway. Portugal, Spain, and Sweden. 19B7 Single European Act The member stales ol the EU declarer] their intention to create a unified market. The Act also amended earlier treaties to ensure further cooperation on foreign policy, 1993 Treaty ün EU The 'Maastricht treaty" extended cOOperetion to poBiical activities, including Irjroign policy, Tins treaty also lilted 1 he remaining restrictions on migration IromSpein and Portugal tD DlherEU countries fSOPEMI, f 999). 1997 Dublin Convention An attempt to harmonka policy by requiring asylum seekers to apply in the first EU country i hoy enter still no* in eliect- 1999 Trebly ol Amsterdam Ptaced Issues rotating to immigration and asylum under the jurisdiciion of the EC and Incorporated the Schengen Accords Into the EU; included an agreement to achieve minimum standards in asylum policies and practices by 2Q04. 1999 European Council meeting: in Tampere Established the need for a common European policy on asylum and immigration and ashed lite European Commission to draw up proposals on asylum, refugees, and Immigration, £000 Nice Treaty This included a Charter ol Fundamental flighis thai says that non-EU nationals with resJtfence or work permits should eventually have the same freedom of movement as EU nationals. 2001 European Council meeting in Laakon Failed lo agree an greater cooperation on immigration or asylum policies. We: 'in the case of treaties this refers lo the year in which they came Into force ataiKer As far as immigration from outside the EU is concerned, governments still prefer national policies to supranational ones and have proved reluctant to transfer authority to European bodies such as the European Parliament or the European Court of Justice (Koslowsky, 1998). This determination to retain sovereign control over immigration was confirmed at a meeting of the European Council (heads of government) meeting at Laeken in December 2001 which, while calling for closer cooperation to protect external frontiers, rejected a proposal to create a common European border patrolled by EU border guards. There was a similar impasse at this meeting when it came to policies on asylum. The UK, for example, was pressing for common standards on accepting asylum applications but Austria and Germany opposed such measures which would probably require them to accept higher numbers. The European Commission does now have more authority to propose laws on immigration and asylum, but still has to present these to the Council of Ministers where they are subject to close national scrutiny. National policies At the national level the current trend seems to be for governments to make a more realistic assessment of the need, both economic and demographic, for immigrants while also trying to manage immigration more efficiently. National level policies on immigration involve efforts to: control immigration, which include measures to tighten up border controls and to simplify and speed up the processes for dealing with asylum applications; combat illegal immigration, which typically include sanctions against airlines or other travel operators, as well as heavy fines for employers of illegal immigrants - a number of countries have also had regularization programmes for current illegal immigrants; and better integrate immigrants, which include, for example, training for local authorities to make them more sensitive to the needs of immigrants, training and language classes for immigrants, systems of sponsorship to help immigrants settle, and special reception classes for children. The following are some recent measures taken by individual countries: Denmark - In 2000 the Government enacted legislation to deter any immigrant younger than 25 from bringing a foreign spouse to Denmark. And in 2001 the new-elected Government introduced measures to make the country less attractive to refugees and immigrants. Now refugees can be sent home up to seven years after being granted asylum if their home countries are by then deemed safe. And they must also wait seven years before being granted permanent residence permits. France - In 1997 the new Government instituted a regularization programme that has granted residence to 75,600 foreigners. Three-quarters of these came Migration trends and migration policy in Europe from Africa, with similar numbers from the Maghreb and sub-Saharan Africa. In 2001 France signed a bilateral agreement with Algeria on the status of Algerians in France. Germany — In 2001 the Minister of the Interior said that "Germany is an immigration country", a significant shift from the previous official stance. Germany is also considering a new immigration bill that would allow highly skilled people to qualify for permanent residence as soon as they arrive, and also introduced measures to integrate foreigners and tighten the asylum system. However the bill has met with some opposition from parties opposed to immigration and may not be accepted by the upper house of parliament. Ireland - The situation here is somewhat different. Ireland, which has only recently become a country of immigration, and had the fastest-growing economy in Europe, was happy to welcome more workers. Here around three-quarters of immigrant workers are unskilled. Now Ireland is becoming more restrictive. Faced with rising unemployment it is tightening up on work permits. Employers now have to show that reasonable efforts have been made to recruit EU nationals. Italy- So far Italy has granted residence permits to foreigners who have entered legally or illegally and found work. Early in 2002, however, there were plans for a tougher immigration law that would permit non-EU immigrants to remain in the country only for as long as they have a contract of employment, and also make it simpler to expel illegal immigrants. Sweden - Sweden has always made efforts to integrate its foreign labour force and has never operated under the assumption that contract workers would go home. Here it is relatively easy to gain citizenship and around half the foreign-born are citizens. In 1998 the Government introduced a new policy on integration and in 2001 enacted a new citizenship law that recognizes dual nationality. United Kingdom - During 2001 British policy on labour immigration shifted significantly in favour of allowing more foreign workers. From the beginning of 2002 a "Highly Skilled Migrant Programme" will use a points scheme based on educational attainment and salary to admit foreign professionals who do not have a pre-arranged job. And the opportunities for "working holiday makers" and seasonal workers have been widened. At the same time the penalty for smuggling or trafficking people has been increased from a ten to 14 year prison sentence (Koslowsky, 1998). INTERNATIONAL COOPERATION Thus far, European countries have pursued most of their migration policies within their own national borders by controlling the entry of foreigners and supervising 78 Stalker the integration of existing immigrants. They have done relatively little to try to control immigration through cooperation with the sending countries. In earlier eras, when the aim was to encourage labour immigration, Germany set up recruitment bureaus in the former Yugoslavia and Turkey, and France and the UK at various times established systems to encourage immigration from their former colonies. Even today, a number of European countries have bilateral quota agreements with sending countries for unskilled workers, usually for temporary or seasonal labour. Germany has the largest number of seasonal workers, most of whom come from Poland, and France has bilateral agreements with Morocco, Poland, Senegal, and Tunisia (IOM, 2000). There are also bilateral arrangements - "readmission agreements" - for the repatriation of various types of migrants, usually failed asylum seekers. Although most such arrangements are bilateral, some are multilateral. Most of the latter are through the International Organization for Migration (IOM), which among other activities, helps people who have to return home - rejected asylum seekers, trafficked migrants, stranded students, and some labour migrants. One project, for example, concerns the "Reintegration and Emigration of Asylum Seekers from Germany" which in 2000 assisted some 70,000 people to return to Bosnia, Kosovo, and elsewhere (IOM, 2002). Aid in place of migration? Could the receiving countries also do more to prevent migration by addressing the causes of unwanted immigration rather than simply trying to control it? At times there have indeed been suggestions that the richer countries might target some of their development aid in this direction, using it to defuse potential conflicts that could trigger flows of refugees, for example, or to alleviate the poverty that causes people to seek work overseas. Conflict prevention and resolution came to be seen as a more urgent task during the 1990s following a surge of internal conflicts in Europe and elsewhere. The European Community, for example, when preparing Country Strategy Papers for the countries to which it gives aid now assesses the potential for conflicts - looking at such issues as the balance of political and economic power, the nature of the security forces, the ethnic composition of the government, the representation of women, and the extent of environmental degradation. There have also been efforts to focus on countries that have produced a large number of emigrants. In 1998 the EU created a High-Level Working Group (HLWG) on Asylum and Migration which has now developed Action Plans for Sri Lanka, Somalia, Albania, Afghanistan, Iraq, and Morocco, aimed at Migration trenas ana migration pouty in curupe comprehensively addressing the roots of displacement (European Commission, 2001). In what are always very complex situations, however, it will be difficult to prove whether such interventions really do have a significant impact on emigration. It is equally difficult to judge the impact on emigration of aid given to reduce poverty. This was an issue raised explicitly at several international conferences in the early 1990s, and still remains largely unresolved (Böhning and Schloeter-Pareses, 1994). Many of the doubts concern whether aid can indeed reduce poverty - a much larger question beyond the scope of this paper. But even if development cooperation did reduce poverty it is questionable whether this would then immediately stem emigration. A number of studies have concluded that when people's incomes and aspirations rise as their countries industrialize they will become more mobile both nationally and internationally and have the resources to emigrate. Only later, when the labour market at home offers sufficient remunerative employment,will the more ambitious people be content to remain at home. This produces what has been called a "migration hump" as migration first rises and then falls (Martin and Taylor, 1996). How rich do people have to be before they do not feel impelled to emigrate? Some studies in the mid-1990s suggested that the transition occurred at an average real per capita income of around $4,000. This is illustrated in Figure 4 which suggests that the transition occurs first for national migration, then for international unskilled migration, and finally for migration of the highly skilled (Fisher and Straubhaar, 1996). Nevertheless it could also be argued that even if this is the case it might be useful to try more targeted interventions to boost employment, specifically in countries and areas that send large numbers of migrants. The evidence here is not very positive. The International Labour Organization (ILO), for example, has been involved with such activities in high-emigration areas in the Maghreb countries, which do have per capita incomes of around $4,000. These have included "micro-level" targeted interventions such as support to small enterprise development. The ILO has concluded, however, that such interventions usually have no perceptible impact on migration pressure and argues that a more fruitful option would be for the richer countries to open their markets so as to enable the poorer migrant-sending countries to pursue the kind of export-led growth that could give a more substantial boost to both wages and employment (Abella, 2002). The empirical evidence Do European countries, in practice, try to direct aid to reduce migration pressures? One indication would be if European ODA flows were concentrated on migrant-sending countries. Table 7 compares, for the countries with both sets of data available, the main developing-country destinations of ODA with the Stalker main developing-country sources of immigrants (DAC, 2001). The developing countries listed here are the leading three, though countries that also figure highly in the top 12 or 15 in the other category are also asterisked. Thus for Belgium, Morocco is the leading source of immigrants, and is also ranked highly as an aid destination (number 15). This table does indeed suggest a degree of correlation. However, in most cases this is because both immigration and aid flows reflect colonial ties. For the UK, India is the leading destination of ODA primarily because of a long historical association, rather than from an attempt to dampen emigration. The strongest colonial correlation is for Portugal, almost all of whose assistance goes to former colonies. Turkey also figures highly on the ODA lists, though again there are probably other reasons for this, since Turkey, particularly during the Cold War, was strategically important to Europe. On the other hand, Tanzania figures strongly as an aid recipient, though it is not a significant source of migrants. The lack of any clear connection between aid policies and migration is also implied from the published policies of donors. The DAC guidelines on poverty reduction, for example, do mention migration but only in the broader context of the ways in which development assistance could contribute to poverty reduction with the implication that this in turn might reduce emigration pressures. THENEXTPHASES OF EUROPEAN IMMIGRATION In the past most European countries have not considered themselves countries of immigration. Their first instinct has been to resist large numbers of new arrivals. Recent developments, however, suggest possible changes of direction. Although in the short term there can be switches in immigration policy in response to immediate political pressures, in the longer term immigration is likely to grow. This is partly because of labour demand, since even at times of slow economic growth, most European countries find themselves short of skilled workers and also of people prepared to do jobs that national workers shun. The longer-term picture will also be affected by demographic changes and the greying of the population. Immigration is not the answer to falling birth rates, since countering this would require immigration on a vast scale. What demographic shifts could eventually do, however, is erode popular resistance to immigration and encourage governments to accept more people, even if in a closely controlled fashion tailored as precisely as possible to national needs. For the EU in the short and medium term many of these labour demands are likely to be met from the East, rather than from the South. As the EU expands eastward it will gain access to new sources of migrant labour, similar to those provided in the 1950s and 1960s by Italy, Portugal, and Spain. Yet, given the low birth rates in most of these countries too, the longer-term picture, even for an expanded EU, is likely to involve greater use of workers from developing countries. TABLE 7 LEADING SOURCES OF FLOWS OF IMMIGRANTS, AND LEADING DEVELOPING-COUNTRY DESTINATIONS OF OFFICIAL DEVELOPMENT ASSISTANCE (ODA), SELECTED COUNTRIES, 1998-1999 Belgium Denmark Finland France Germany Greece | taty Netherlands Norway Portugal Sweden Switzerland UK Top Throe Soutces of Migrants. 1999 Morocco" Turkey Iraq Somalia Iraq* Lraii Malawi Turkey China" Morocco" Algeria' Turkey Turkey ■ - Egypt" Philippines Turkey' Morocco China" Senegal Morocco Turkey Suriname' Iraq" Somalia trän Brazil * Guinea-Bissau' Cape Verde- trait han China Turkey - - S- Africa* India' Pakistar Top Three Destinations of ODA, 1993-1933 Tanzania Congo Tanzania Uganda Rwanda Mozambique Mozambique Nicaragua China' Cůleďlvoire Ff. New Polynesia Caledonia China Indonesia Turkey* Palestino Egypl* Turkey" Madagascar Mozambique H alii Neth, Tanzania india AnUlles Tanzania Mozambique Palestine Mozam- East Timor Cape Veide' bique" Tanzania Mozambique S. Alrica Bangladesh Mozambique India India* Tanzania Bangladesh 4otss: *Also a leading source of migrants, or also a leading destination of ODA, though not necessarily m lha top ihree. This does not include European counuies such as Bosnia and Herzegovina and the lorrner Yugoslavia, which &re major immigrant sources, as well as destinations far ODA. Sources: SOPEMI. 2001 i Development Assistance Committee, 2000. NOTE The European Economic Area includes the 15 members of the European Union plus Iceland, Liechtenstein, and Norway. REFERENCES Abella, M. 2002 Personal communication with Manolo Abel la, Chief of the ILO International Migration Branch. Böhning, W., and M. Schloeter-Paredes 1994 Aid in Place of Migration?, International Labour Organization, Geneva. Borrie, W. 1970 The Growth and Control of World Population, Weidenfeld and Nicolson, London: 94. DAC 2001 The DAC Journal: Development Assistance, 2000 Report, OECD, Paris. DRC 2001 The Dublin Convention: Study on its Implementation in the 15 Member States of the European Union, Danish Refugee Council/European Commission, Copenhagen. European Commission 2001 Communication from the Commission on Conflict Prevention, European Commission, Brussels. Eurostat 2000 European Social Statistics: Migration, Eurostat, Luxembourg. 2002 Collection: key indicators; Theme: Population and Social Conditions, Eurostat website, http://europa.eu.int. Fischer, P., and T. Straubhaar 1996 "Is Migration into EU countries demand based?", in D. Corry (Ed.), Economics and European Union Migration Policy, Institute for Public Policy Research, London. Hollifield,J. 1992 Immigrants, Markets and States: The Political Economy of Postwar Europe, Harvard University Press, Cambridge. International Organization for Migration (IOM) 2000 World Migration Report, IOM, Geneva. 2002 http://www.iom.int/en/what/main_heading_reag.shtmI. Koslowsky, R. 1998 "European Union migration regimes, established and emergent", in C. Joppke (Ed.) Challenge to the Nation-state: Immigration in Western Europe and the United States, Oxford University Press, Oxford. Martin, P., and J. Taylor 1996 "The anatomy of a migration hump", Development Strategies, Employment and Migration: Insights from Models, OECD, Paris.