There is hardly another region of the worldwhere the current situation of migrations isstill considerably influenced by the past his-tory as in the Balkans. Migrations have beena fundamental element in the history of theBalkans, accompanying its stormy events[Her√ak and Mesic, 1990] and obviouslycontinuing to do so, even at the start of thenew millennium. For centuries, invasions,wars, military defeats and victories havebeen a more or less direct cause of popula-tion movements, in a continuous and stillongoing transformation of the distributionand the overlapping of religions, languages,ethnic groups and cultures [Sardon, 2001].Since the arrival of the Slav populations inthe 7th century, the Ottoman expansion, theextension of the Hapsburg domain, the riseand growth of national states, the two WorldWars and the ethnic wars of the past decade,there have been numerous population move-ments caused more or less directly by polit-ical and military events.

In some cases, the legacy of migrations thattook place centuries ago is still tragicallyrelevant, and weighs on the destiny of theBalkan peoples. The forced migrations fromKosovo or Krajina are the most significantexamples, although certainly not the onlyones. Long-term historical factors still seemto enhance the specific and particular

aspects of the Balkans [Prévélakis, 1994],affecting their specific nature also withregard to migration. However, the migrationoutlook of the Balkans does not just involvethese flows, which in some respects tend toreflect the movements de l’histoire delongue durée, but it is rather much morestructured and complex. Focusing for themoment on the most recent period, togetherwith the forced migrations caused by theethnic conflicts in the former Yugoslaviaand the ethnic migrations followed by thecollapse of the regimes created by “realsocialism”, we find both forms of labourmigration and transit migration. Therefore,the Balkans are also characterised by manyof the typical elements of the current formsof mobility in Eastern and Central Europe[Okólski, 1998; Bonifazi, 2003].

With regard to the positioning of the individ-ual countries in continental migration, in ashort period of time the situation in theBalkans has undergone extraordinarychanges. As late as the second half of the1980s, the Balkan situation was charac-terised by the presence of two traditionalsending countries, such as Yugoslavia andTurkey, together with a country which waschanging into a receiving country, such asGreece, and other three countries that weresubstantially detached from the European

ESPACE, POPULATIONS, SOCIETES, 2004-3 pp. 519-531

Corrado BONIFAZIMarija MAMOLO

Istituto di Ricerche sulla Popolazione e le PoliticheSociali IRPPSvia Nizza, 128RomaItaliec.bonifazi@irpps.cnr.it

Past and Current Trends ofBalkan Migrations

INTRODUCTION

labour-migration flows, such as Albania,Romania and Bulgaria. Currently, however,these latter three countries should be consid-ered as being among the most importantsources of European emigration. Turkey hassubstantially maintained its role, although itsrange of destinations is much broader.Greece has definitely become a receivingcountry, while most of the former Yugoslaviais still far from having reached a completelynormal condition in the migration field.This paper will seek to provide a generaloverview of developments in international

migrations in the Balkans as from the end ofWorld War II. This survey will regard theBalkan countries that have experienced“real socialism”, and where, for obviousreasons, the most radical and interestingchanges and innovations have occurred.The purpose is above all to highlight theinternal differences in the Balkans andthose between the Balkan countries and therest of the continent, also taking into dueaccount the recent context characteristicsbriefly examined in the last section of thispaper.

520

1. BALKAN MIGRATION IN THE DECADES OF THE COLD WAR

By and large, the statistical measurement ofinternational migration is one of the most dif-ficult tasks in the domain of demographicdata collection. Limitations and difficulties ofmeasuring spatial movements of populationare well known. Besides, concepts, defini-tions and data collection systems differ acrosscountries and times, making spatial and tem-poral comparisons a really fuzzy task. Wars,conflicts, economic and social turmoil con-tribute to largely increase the difficulties inmigration statistics collection, reducing theirquality and affordability. Last but not least, inmany cases the lack of flow data compels torefer to stock data. Notwithstanding theselimitations, available data offer interestinginformation on Balkan migration when prop-er caveats are considered. The international migrations involving thecountries with planned economies in theBalkan peninsula from the period afterWorld War II up to the 1990s can be attrib-uted to two main types of flows: ethnic andlabour migrations. The former concernedthe German, Turkish, Hungarian and SlavMoslem minorities settled in several Balkancountries, the Italians of Yugoslavia, theJews as well as Gypsies of Romania (Table1). Labour migrations, on the other hand,have exclusively regarded Yugoslavia.With the end of the war and the peace treaties,360,000 Germans left Yugoslavia forGermany and Austria. Some of theHungarians from Vojvodina and Romaniamoved to Hungary and 200,000 Italians left

the territories ceded to Yugoslavia [Fassmannand Münz, 1995]. These are, of course,approximate estimates, not at all precise andrigorous, and above all agreement does notalways exist on these figures. For example, inthe case of the Italians, other sources consid-er 250,000 to be a more realistic estimate ofthe size of the migration flow [Donato, 2001].The data starting from 1954 on the flow ofthe Turks from Yugoslavia to Turkey are con-sidered to be more accurate. The creation of asocialist State was not welcomed neither bythe population of Turkish origin nor by theMoslems in general, so that after relationswere restored between the two countries, apopulation flow started, involving 151,000people between 1954 and 1960, with 30,000in the subsequent decade and just over 4,000between 1970 and 1990 [Kiris

5çi, 1995].

These figures, besides the Turks, also includemany Albanians and Moslems from Bosniaand Sandjak who declared a different nation-al origin in order to be able to emigrate toTurkey.The ethnic migrations from the Balkans toGermany did not end in the years immedi-ately after the conclusion of the war. In theperiod surveyed, a large number of people ofGerman origin emigrated from Romania. Inparticular, starting from the 1970s, Germanypromoted the emigration of Romanian citi-zens belonging to the German minority as aresult of bilateral agreements signed withthe Romanian government. From then up to1989, approximately 230,000 people of

German origin emigrated from Romania[Fassmann and Münz, 1995]. Between 1960and 1992, most of the Romanian Jews alsoleft the country, moving mainly to Israel andthe United States. Furthermore, after 1987approximately 60,000 Hungarians leftRomania legally, most of them settled inHungary [Dövényi and Vukovich, 1994].With regard to information on other minori-ties, in the period 1950-53 approximately250,000 Turks obtained permission to emi-grate from Bulgaria, and later, on the basisof the 1968 agreements, over 95,000 Turkswere granted the right to emigrate to Turkey.Out of the latter, only 14.1% decided toremain there [Bobeva, 1994].With regard to Yugoslavia, starting from the1960s, migrations were also due to strongeconomic motives. Yugoslavia was the onlycountry with a socialist regime, not only inthe Balkan Peninsula but in the whole ofEast-Central Europe, to tolerate and evenpromote this type of emigration. Betweenthe 1960s and 1970s, the country becameone of the major sources of labour force,thus contributing to the industrial develop-ment of North-Central Europe. Germany,Switzerland and Austria were the main hostcountries. Yugoslav researchers estimate acontinuous growth in the stock of migrant

workers living in Western Europe between1964 and 1973, the year when the figure of850,000 people should be recorded[Malaçic, 1994]. The policies to stop immi-gration, introduced in the major receivingcountries, led to a definite reversal in trends,stopping only in 1985 when labour emigra-tion stock was reduced to 500,000 people. Inthe last five years of the 1980s, the valuesgrew once more, also due to the effect of theprofound economic, social and political cri-sis under way in Yugoslavia, with a stock of550,000 emigrant workers being estimatedin 1990 [Malaçic, 1994]. The statisticalsources of the most important Europeanreceiving countries generally refer to theforeign population, considering togetherworkers and inactive population. Accordingto these data, in the early 1990s the overallnumber of Yugoslavian labour migrants andtheir family members exceeded 1.3 millionpeople [Fassmann and Münz, 1995]. A largepart of the discrepancy between these twoestimates of Yugoslavian emigration is theresult of the different groups of populationconsidered (workers vs. workers and familymembers), but it also reflects the differentinterest and ability of receiving and depar-ture countries in measuring the two sides ofthe same migration flow.

521

Table 1. Ethnic international migration in some Balkan countries in the period 1945-1992/93, absolute values

Source: Fassman and Münz (1995).

For the Balkans, the fall of the Berlin Walland the conclusion of the cold war repre-sented the end of the longest period of sta-bility and peace in their recent history[Prévélakis, 1994]. In the region, the transi-tion process had two main types of conse-quences on the migration dynamics. Firstly,by triggering the ethnic conflicts it led to theconditions producing forced migrationsfrom the countries of the former Yugoslavia,which represent undoubtedly the most dra-matic and characteristic element of all theEuropean migrations in the last decade.Secondly, it created the political and eco-nomic basis (as well as in the rest of East-Central Europe) for the continuation or start-up of a series of migration flows that werepreviously strictly controlled by the State.Therefore, since 1989 there has been a fur-ther revival of the ethnic migrations alreadyconsidered in the previous paragraph andlabour migration flows from countries thatwere previously totally excluded from thesedynamics developed as well. Finally, mostof the Balkan region has become highlycharacterised as a migration transit zone, asa place of origin, as well as a land of passagefor various forms of illegal migration[Bonifazi, Conti and Mamolo, 2003].The quantification of these processes is farfrom being satisfactory. In fact, in additionto the already mentioned problems in mea-

suring international migration specific fac-tors have contributed to make even moredifficult the data collection in the Balkanarea in the last fifteen years. Ethnic wars informer Yugoslavia have been a powerfulfactor in reducing, directly and indirectly,the availability and the quality of migrationstatistics. Besides, as in all Eastern andCentral Europe, the economic and socialproblems of the transition period have deter-mined a climate not particularly favourableto the organisation and implementation of acomplex recording system for a phenome-non that up to a few years ago hardly exist-ed. Finally, other problems arise from thenew characteristics of migration processes,with the increasing importance of informal,irregular or illegal flows. Thus, the availablestatistics describe, though not always fully,the legal component of migration, providinglimited data on the flows going along differ-ent paths.In order to have a general idea of the extentof the migrations, reference has been madeto two different estimates of net migrationsin the countries considered in the 1990s(Table 2). The first estimate is based onmigration statistics, in particular those con-cerning the receiving countries, and the sec-ond estimate on the calculation of the migra-tion balance as the difference between thetotal growth of the population and the natur-

522

2. SIZE ESTIMATIONS OF MIGRATION SINCE THE END OF THE COLD WAR

Table 2. Net migration levels in Balkan countries, 1990-2000 (a)(annual rates per thousand, annual absolute values in thousands)

Source: UN (2002a and 2002b).Notes: (a) In the first two columns, net migrations are based on the official migration statistics. The last five columns

show the residual of the demographic balance equation (total population growth minus natural growth).

al balance. Both values are calculated by theUnited Nations and highlight considerabledifferences, which are in any case also use-ful to illustrate the main trends.The highest migration loss of the decade isthe one recorded in Bosnia between 1990and 1995, with an annual rate of -43.2 per1000 according to the migration statisticsand -52 per 1000 according to the estimatesof the net migration as residual. Accordingto the latter source in absolute terms, thetotal loss would have been one million peo-ple in a country with a population of just4.3 million at the start of the decade. Againaccording to this measurement of themigration balance, half of this loss wouldhave been recovered in the subsequent 5-year period. In fact, from 1996 the processof return from asylum countries started,leading after 1997 to the returns exceedingthe departures [UN, 2002a]. The sharp dif-ference in the period 1995-98 between thevalues shown in the two different methodsof calculation induces us to consider theinformation available with great caution,until more reliable measurements areobtained.Although Bosnia recorded the highest levelsof migration loss, the major sending countryof the region seems to be Albania. Also inthis case, the differences in the measurementof the net migration are considerable, andshould mainly be attributed to the high num-ber of undocumented migrants moving toItaly and above all to Greece. According tothe statistics of the receiving countries, theaverage annual rates are -2.5 per 1000 and -8.4 per 1000, while in the estimates usingthe residual method the figures are -24.7 per1000 and –21 per 1000. The latter valueswould signify a total migration loss of

700,000 people in the 1990s. This figure, aswe shall discuss later, is in basic agreementwith the most recent evaluations of Albanianimmigration formulated by the officialGreek and Italian sources.Bulgaria and Romania are two other sendingcountries although at much lower levels.The data available show, in the two 5-yearperiods examined, a constant loss forBulgaria and a declining loss for Romania.In the former case, the migration loss wouldhave been 40,000 people per year, and in thelatter it would have fallen from 106,000 to12,000 people. In both countries, emigrationin the early 1990s was strongly ethnic incharacter, though in subsequent years ittended more towards labour migrations.With regard to the other countries of the for-mer Yugoslavia, Slovenia and Croatia recorda positive migration balance, whileMacedonia and Yugoslavia reveal an alter-nation between positive and negative bal-ances. Slovenia, which entered theEuropean Union in May 2004, highlights thebest economic performance among thecountries surveyed, and the positive bal-ances recorded are most probably the signsof a coming transition of its role in theEuropean migrations, with the transforma-tion into a receiving country. For Croatia,this type of role still seems far away, and thepositive balances recorded are related to thedynamics of forced migrations caused by theconflicts. The same reasons have also con-siderably influenced the migration dynamicsin Serbia and Montenegro, with a highlypotential underestimation of the arrivals ofdisplaced people, also leading to contrastingevaluations of net migration rates. Migrationhas been modest in Macedonia and is alsolargely due to conflicts.

523

3. FORCED MIGRATIONS AND STOCK OF BALKAN EMIGRANTS INWESTERN EUROPE

Moving on from an estimate of migrationbalances to a complete reconstruction of thevarious flows characterising the countriesconsidered in the past 15 years seems to bea rather difficult task. In a number of cases,in fact, the measurement of Balkan migra-

tions appears to be insufficient, inaccurateor wholly lacking. If, however, we refer tothe number of displaced people hosted in acountry of the former Yugoslavia and to thenumber of citizens of a Balkan country resi-dent in a country of Western Europe, these

two sets of stock data can be used to deter-mine some important characteristics of themain migration flows of the period.The greatest flows of displaced people, withregard to those remaining within the bordersof the old socialist Yugoslavia, moved fromCroatia and Bosnia to Yugoslavia and fromBosnia to Croatia (Table 3). The stock dataavailable, despite all the limits, show threemigration streams, which, at their maximumpeak, have led to a displaced populationwith a similar size. The values reached287,000 people in 1993 for Bosnian dis-placed persons to Croatia, 349,000 in 1992for the Bosnians hosted in Yugoslavia and299,000 in 1999 for the forced migrationfrom Croatia to Yugoslavia. On the whole,the data follow the trends of the conflictsand reflect the persistence of certain situa-tions being still far from a satisfactory solu-tion, especially with regard to the Croatianand Bosnian Serbs displaced in Serbia andMontenegro. Despite their dramatic nature,the data shown in Table 3 only partiallyreflect the intensity of forced migrations inthe former Yugoslavia. To these figures weshould add the internally displaced personsand flows directed beyond the borders of the

former Yugoslavia. These flows were espe-cially heavy during the most violent phasesof the conflicts and stopped only when mostof the host countries significantly restrictedentries to asylum seekers. For example, thenumber of internally displaced persons atthe end of 2002 totalled 367,000 in Bosniaand 262,000 in Serbia and Montenegro, fig-ures to which we should add some 85,000people under the protection of the UNHCRin Kosovo. Between 1992 and 2002, thetotal number of asylum applications lodgedin industrialised countries exceeded onemillion, of whom 850,000 from people orig-inating in what is today the Federation ofSerbia and Montenegro and 194,000 frompeople coming from Bosnia-Herzegovina[UNHCR, 2003].In the early 1990s, the number of immigrantscoming from the Balkans residing in a coun-try of Western Europe was especially high inGermany, Austria and Switzerland andalmost exclusively involved the Yugoslavs(Table 4). In 1991, they reached a total of663,000 people in Germany, 198,000 inAustria and 141,000 in Switzerland. In rela-tive terms, and also considering the otherBalkan countries, migration from this area

524

Table 3. Refugees in former Yugoslavia countries by country of origin and asylum, 1992-2002.Absolute values in thousands

.. not reported by the sourceSource: UNHCR (2002, 2003)

represented 12.8% of the entire foreign pop-ulation in Switzerland, 13.8 in Germany and42.7 in Austria. In subsequent years, thesecountries have continued to be three of themain destinations of Balkan emigration. InGermany, the foreigners from the Balkansexceeded 1.5 million people in 1996, i.e.21% of the total, falling to 1.2 million in2001 due to the effect of the return of manydisplaced persons; in Switzerland, theytotalled 350,000 in 2001 and accounted for aquarter of the entire foreign population. InAustria, they totalled 342,000 in 2001 andaccounted for 45% of the total. Most of thesefigures still refer to immigrants coming from

the countries of the former Yugoslavia; onlyin Germany, a considerable Romanian com-munity has settled down totalling approxi-mately 90,000 people in 2001.

In the 1990s and in the early part of thisdecade, Greece and Italy have emerged astwo other major countries of destination ofthe Balkan emigration, especially fromAlbania. According to the data of the lastGreek census, all the Albanians resident inGreece totalled 438,000 in 2001, to whichwe should add 35,000 Bulgarians and 22,000Romanians, while there were practically noimmigrants from the countries of the former

525

Table 4. Stock of foreign population from the Balkan countries in some European countries, 1st January.Absolute values in thousands

.. not available; - less than 500.Source: Eurostat (2003)Notes: (a) Data refer to former Yugoslavia. (b) Data from OECD (2003). (c) Our estimate. (d) Data from OECD

(1995). (e) Census data from National Statistical Office. (f) ISTAT data.

In the 1990s, various socio-economic anddemographic factors regarding Balkanmigration have emerged, adding new signif-icance to population movements in the area.These elements have not always been con-sidered sufficiently important, with attentionbeing focused on the impact of the forcedmigrations caused by the conflicts. It seemsthat the development of migration flows hasalso followed the guiding lines of consoli-dated historical and cultural links and themore recent trade flows between the Balkansending countries and the European receiv-ing countries. All of these factors as a wholehave become increasingly important in

guiding migration flows in the region. Thissection will focus on some of these contextfactors in order to seek to provide some ele-ments to think of when exploring the presentand the future prospects of the migratoryprocess in the region.At the start of the current decade, theamount of the population of the 8 Balkancountries considered was approximately15% of the population of the 15 EU MemberStates (Table 5). Romania showed thelargest population with approximately 22.5million, while Macedonia and Slovenia hadthe smallest populations, with approximate-ly 2 million citizens. In the 1990s, some

Yugoslavia. Overall, the Balkan componentrepresented 65.6% of the total immigrationand the Albanians alone formed 4% of thetotal population of Greece. In Italy, duringearly 2001, the resident Balkan immigrantstotalled 352,000, i.e. 24% of the foreign pop-ulation. Also in this case, the Albanians werethe largest community, tending towards agradual stabilisation [Bonifazi and Sabatino,2003]. The Romanian component was grow-ing: this trend is also confirmed by data ofthe last regularisation, since out of the700,000 applications Romanian citizens sub-mitted 143,000. The number of applicationssubmitted by the other Balkan nationalitieswas 54,000 for the Albanians, 9,100 for theBulgarians, 6,700 for the Yugoslavs, 5,800for the Macedonians, 4,200 for theCroatians, 2,700 for the Bosnians and 500for the Slovenians.Also in some countries of Northern Europe,the percentage of immigrants from theBalkans exceeded 10% of total immigrationin the 1990s. In absolute terms, these valuesexceeded 23,000 people in Norway and36,000 in Denmark in 2001, while the fig-ures for Sweden in 1996 exceeded 100,000.In the latter country, the fall between 1996and 2001 is due to an increase in the numberof naturalisations, totalling 35,800 for citi-zens of the former Yugoslavia in the 2-yearperiod 1998-99 [Bonifazi, Conti andMamolo, 2003].

The data considered reflect only the regularcomponent of migration from the Balkanarea to Western Europe, whilst the size ofirregular or illegal migration can be relevantfor some flows. The same magnitude of reg-ularisation processes shows the gap existingbetween the official and the actual size ofinternational migration. In this respect, datafor Romanians in the last Italian amnestyare really significant. They give evidencefor the presence of an important flow oflabour migration in the last years, whileaccording to UN estimations in table 2 netemigration of Romania largely decreased inthe second half of the nineties. It is possiblethat trends in Romanian migration haveabruptly reversed at the turn of the centurywith the emerging of new destinations, as itis possible that UN fails to capture allmigration flows. Waiting for new andupdated data, the impression is of arenewed importance of labour migrationfrom Balkans along different and in manycases informal paths. These developmentsare a sign of the more general evolution ofinternational migration in Central andEastern Europe, with the beginning of newforms of mobility characterised by short-term movements [Okolski, 1998]. This situ-ation can create a sound and large base formore traditional migration movements ifnew channels of entry and stronger pull fac-tors should appear.

526

4. CONTEXT FACTORS AND MIGRATIONS IN THE BALKANS IN THE 1990S

527

Balkan countries experienced a significantfall in the population, generally blamed onmigration losses, which in some cases alsoadded to the natural negative balance. Theeconomic and social crisis of the 1990s andthe military conflicts also determined strongrepercussions on the volume of the working-age population. In the 1990s, the averageannual growth rate was negative in Albania,Bosnia, Bulgaria and Croatia. Whereas, thepopulation aged between 15 and 64increased in all the other countries. Thistrend is basically due to the migration

dynamics characterising these countries,rather than to natural population dynamics.In this respect, Albanian age structureappears the most affected by emigration,when we consider the very high potential ofdemographic growth of the country. As anexample, 600,000 Albanian immigrantswere officially recorded in Greece and Italyat the beginning of the present decade. Thisvalue represents something more than onesixth of the whole population of the country;it mainly regards the working age popula-tion and can heavily affect the age structure.

Table 5. Some basic demographic and economic indicators for the Balkan countries, 2000

Source: (1) Council of Europe (2002); (2) our elaboration on data from UN (2003);(3) World Bank (2003); (4) ILO (2002).Albania: UNDP (2000), Bosnia Herz.: UNDP (2003); EU 15: Eurostat (2002)

Notes: (a) 1999; (b) 2001; (c) Work applicants.

At the end of a decade of crisis, there stillexist some characteristics of the populationdynamics and structure that could have aneffect on the migratory process in the nearfuture. First of all, the natural populationgain is especially significant in Albania,which in 2000 had a natural growth rate of12.2 per 1000. At the end of the 1990s,births also exceeded the number of deaths inBosnia and Macedonia, with a naturalgrowth rate of 3.1 and 6 per 1000 respec-tively, while in the other countries this rate isnegative or almost zero. The natural dynam-ics are also reflected in the differencesobserved in the age structure of the popula-tion. The Albanian population shows a highpercentage of young people aged under 15,who represent almost one third of the totalpopulation, while people aged 65 and above

were about 6% at the end of the 1990s. InBosnia and in Macedonia, the percentage ofyoung people was about 20% in 2000, andthe older population represented approxi-mately 10%. It should in any case be point-ed out that the ageing process has now alsostarted in these countries, as observed forsome time in Bulgaria, Croatia, Romaniaand Slovenia. This involves a fall in the per-centage of the population aged under 15years and an increase of the older popula-tion. A similar situation can be observed inYugoslavia, though with sharp differenceson the regional level. Furthermore, againaccording to the UN estimates for the 1990s,in most of the countries surveyed the rela-tive weight of the working age populationincreased, involving a significant fall in thetotal dependency ratio. However, the posi-

528

tive growth rate for the period 2000-2010 ofthe working age population in Albania,Bosnia, Macedonia and Yugoslavia couldaffect the migration potential, whichalthough low in absolute terms, could stillaffect the neighbouring countries or the tra-ditional European host countries.The purely demographic characteristics ofthe population in the countries of theBalkan Peninsula can in any case only part-ly justify the dynamics in population move-ments of the Balkans. The effect of popula-tion pressure is more obvious when com-bined with other economic and social fac-tors, which could have triggered migrationflows, without counting the forced migra-tions with an evident ethnic nature due tomilitary conflicts.One of the potential factors “triggering”migrations could lie in the increase ofunemployment, officially inexistent duringthe socialist period. Statistics on unemploy-ment trends in 1990s are incomplete and inany case not very reliable. For indicationson the unemployment levels, reference willbe made here to the ILO statistics based onlabour force sample surveys, and in case ofmissing data, to figures provided in therespective national Human DevelopmentReports published by the UNDP. In most ofthe countries considered the total unem-ployment rate is considerably higher thanthe EU 15 average. Furthermore, it has to benoted that the figures reported might hidean even more critical unemployment situa-tion in many of these countries. The case,for example, of people with highly insecureand unstable jobs is disregarded. In 2000,the majority of the countries consideredshow a total unemployment rate above 16%with significant peak for Macedonia, whichrecorded an unemployment rate of 30.5%(Table 5). According to the ILO statisticsrelying on labour force surveys estimates,Slovenia and Romania are the exceptions.Nevertheless, the figures should be takenwith caution as required for the otherBalkan countries. Referring to registeredunemployment, Romania records a totalunemployment rate of 10.5% in 2000.Albania registered a fall in unemploymentrate after the first years of the economictransition. Nevertheless, the favourable sit-uation was reversed after the financial cri-

sis occurred in 1997, with unemploymentrates again rising to 18.2%, thereforeinducing a serious problem for the coun-try’s economic and social system. It shouldfurthermore be stressed that an improve-ment in the situation of the Albanian labourmarket has also been compromised by theextent of the underground economy and thediscrepancies between real and officiallyrecorded unemployment [UNDP, 2000]. InBosnia-Herzegovina, the unemploymentrate in 2001 was estimated at 16.4%. In thissituation, the problems encountered whenwe seek to quantify the level of unemploy-ment are similar to those observed inAlbania, and in particular regard the factthat much employment is covered by theinformal sector. Nevertheless, a large por-tion of actual unemployment is also “statis-tically hidden” within registered employ-ment. Besides the unemployment data, theinstability of the labour market and the eco-nomic insecurity experienced by most ofthe population are also rather worrying[UNDP, 2003].The economic recession in the 1990s alsoemerges in the Gross Domestic Product(GDP) levels. According to the figuresavailable from the World Bank (2003), atthe end of the last decade most of the coun-tries surveyed recorded a lower GDP than in1990. Other indicators, such as the GDPper-capita, also confirm a generally criticalsituation. For most of these countries, thedegree of economic development - if deter-mined on the GDP basis - is currently farunder the European Union average.Apart from the purely economic signifi-cance of trade between the Balkan countriesand the other European countries, the com-mercial link is also important in the contextof population movements considered as anintegral part of the overall factors contribut-ing to the definition of a migration system.In fact, international migration is increas-ingly the result of the very complex set ofrelations linking the different part of theworld under many respects. The globalisa-tion processes have contributed to increasethe importance of the flows of goods, capi-tals and information in establishing, pro-moting and maintaining migration move-ments. In this context, trade flows allow tohighlight the size and direction of the eco-

529

nomic links between Balkan countries andthe other countries, describing an importantstructural factor of international migration.Trade with the countries of the EuropeanUnion is an important segment of the for-eign trade of the Balkan region (Table 6).According to the data for 2000, importsfrom the EU represent the highest percent-age of imports to Albania and Slovenia, and

represent over 50% of the imports in theother countries, except for Bosnia-Herzegovina, Bulgaria and Macedonia.Bosnia-Herzegovina and Macedonia,together with Yugoslavia, are the countriesshowing a higher percentage of importsfrom the other countries of the Balkanregion, while for Bulgaria imports fromRussia total 25%.

The structure of imports according to thecountry of origin suggests the existence ofother types of links between the countries,not necessarily or not only related to eco-nomic factors, but also to geographical, his-torical and cultural aspects. A total of 61%of imports to Albania come from Greeceand Italy. The Italian import rate is approx-imately 19% in Romania and 16% inYugoslavia. In Bosnia-Herzegovina,Croatia and Yugoslavia about 20% ofimports come from Austria and Germany,and this percentage rises to 27% in

Slovenia. Exports follow similar trends. In2000, exports to the EU countries exceeded60% in most of the countries surveyed, witha peak of 89% for Albania. Italy is the mainexporting destination for Albania andmakes purchases between 20% and 30% ofthe exports from Bosnia, Croatia, Romaniaand Yugoslavia. Exports to Germany total27% for Slovenia and range between 10%and 20% for the other countries (except forAlbania and Bulgaria). In any case, weshould also recall the trade flows within theBalkans, with Bosnia, Croatia and

Table 6. Imports and exports in the Balkan region. 2000

.. not availableSource: IMF (2001).Notes: (a) Included only the eight specified countries; when data not available for a specific flow,

this value is considered zero.

530

Macedonia partly reflecting the continuityof the economic links in the old YugoslavFederation. It is interesting to note thatsome of the relations in migration field aresubstantially confirmed considering inter-national trade. Albania’s trade is almostcompletely linked to Italy and Greece,

while Germany has strong ties with coun-tries of Former Yugoslavia, and Romaniaappears to prefer Germany and Italy. In gen-eral terms, economic links expressed byinternational trade might constitute animportant component of contextual factorsof migration trends.

CONCLUSIONS

In the light of what has been observed in thelast decade and of the more recent situation,the Balkan region will obviously maintain asignificant role in the migration dynamics ofthe coming years. There are various factorsthat could contribute to the further develop-ment of the mobility of the population in theregion. The most obvious factors regard theserious economic imbalance between themajority of the Balkan countries and thenearby countries of the European Union,which represent an important point of attrac-tion. We should therefore recall that theworst negative effects of the transitionprocess towards market economies inCentral Eastern Europe have been recordedin the Balkan countries, with the sole excep-tion of Slovenia [UN, 2002a]. To these ele-ments, we should add both an unemploy-ment rate that will be hard to reduce in theshort term and, in some of these countries, apersisting population pressure on the labourmarket due to the still significant potentialof natural population growth.We should furthermore consider that theBalkan region, despite the efforts and com-mitment of the international community, isstill characterised by serious political insta-

bility. This situation seems to have had a larg-er influence on the development of transitroutes for illegal migration than on the size ofthese flows [Bonifazi and Strozza, 2004]. Tothis regard, as previously mentioned in two ofthe EU countries neighbouring the formerYugoslavia, i.e. Greece and Italy, the weightin recent immigration flows of the countriesmost involved in the conflicts has been defi-nitely modest. The Balkans are on two of themain transit routes for illegal migrations, onemoving from the Ukraine towards the CzechRepublic and Slovakia, and the other directlyinvolving Romania, Bulgaria, the formerYugoslavia and Albania [IOM, 2000]. Thetransition period in the region, even on theinstitutional level, has contributed to the pro-liferation of new types of “illegal migra-tions” in the region, such as trafficking andsmuggling, mainly towards the EU coun-tries. On the whole, in the present situation itis hard to forecast the outcome, which close-ly depends on the stabilisation processesunder way. The future migration dynamics inthe most unstable area of the Balkans dependon the outcomes of the more general politicaland social development of the countries ofthe region.

531

REFERENCES

BOBEVA D. (1994), “Emigration from and immigra-tion to Bulgaria”, in H. Fassmann and R Münz (eds.),European Migration in the Late 20th Century,Aldershot, Edwar Elgar.

BONIFAZI C. (2003), “Central and Eastern Europeancountries and the new reality of European internationalmigration”, in I.E. Kotowska and J. Józwiak (eds.),Population of Central and Eastern Europe. Challengesand opportunities, Warsaw, Statistical PublishingEstablishment.

BONIFAZI C., C. CONTI and M. MAMOLO (2003),Le migrazioni internazionali nei Balcani durante glianni novanta, paper presented at the Giornate di studiosulla popolazione, Bari, 27-29 January.

BONIFAZI C. and D. SABATINO (2003), Albanianmigration to Italy: what official data and survey resultscan reveal, Journal of Ethnic and Migration Studies,vol. 29, n° 6.

BONIFAZI C. and S. STROZZA (2004),L’immigrazione nell’Unione europea tra problemi dimetodo e tendenze recenti, Salute e Società, n° 2.

Council of Europe (2002), Recent DemographicDevelopments in Europe 2002, Strasbourg, Council ofEurope.

DONATO C. (2001), “Il problema della quantifi-cazione”, in C. Donato (a cura di), Spostamenti di popo-lazione e trasformazioni sociali nella provincia di Triestee nel distretto di Capodistria nel secondo dopoguerra,Trieste, Regione autonoma Friuli Venezia Giulia.

DÖVÉNYI Z. and G. VUKOVICH (1994), “Hungaryand the International Migration”, in H. Fassmann andR. Münz (eds.), European Migration in the Late 20th

Century, Aldershot, Edwar Elgar.

Eurostat (2002), Economic Portrait of the EuropeanUnion 2002, Luxembourg, European Communities;available online at: http://europa.eu.int, December2003.

Eurostat (2003), New Cronos Database Cd Rom,Luxembourg, Eurostat; Extraction date: February 2003.

FASSMANN H. and R. MÜNZ (1994) (eds.),European Migration in the Late 20th Century,Aldershot, Edwar Elgar.

FASSMANN H. and R. MÜNZ (1995), “European EastWest migration”, in R. Cohen (eds.), The CambridgeSurvey of world migration, Cambridge, CambridgeUniversity Press.

HER◊AK E. and M. MESIC (1990), L’espace migra-toire de Yougoslavie: historique des migrationsyougoslaves, Revue européenne des migrations inter-nationales, vol. 6, n° 2.

International Labour Organization – ILO (2002),Yearbook of Labour Statistics, Geneva, ILO.

International Monetary Fund – IMF (2001), Directionof Trade Statistics, Washington D.C., IMF.

International Organization for Migration - IOM (2000),Migrant trafficking and human smuggling in Europe: areview of the evidence, Geneva, UN.

KIRIS5ÇI K. (1995), Post Second World War immigra-

tion from Balkan countries to Turkey, New perspec-tives on Turkey, n° 12.

MALAÇIC J. (1994), “International EconomicMigration in Former Yugoslavia since 1960”, in H.Fassmann and R. Münz (eds.), European Migration inthe Late 20th Century, Aldershot, Edwar Elgar.

OECD (various years), Sopemi Trends in InternationalMigration. Annual Report, Paris, OECD.

OKÓLSKI M. (1998), Regional Dimension ofInternational Migration in Central and Eastern Europe,Genus, vol. 44, n° 1-2.

PRÉVÉLAKIS G. (1994), Les Balkans. Cultures etgéopolitique, Paris, Nathan.

SARDON J.-P. (2001), Demographic change in theBalkans since the end of the 1980s, Population: anEnglish Selection, vol. 13, n° 2.

United Nations (2002a), International migration fromcountries with economies in transition: 1980-1999,New York, United Nations.

United Nations (2002b), International migration report2002, New York, United Nations.

United Nations (2003), World Population Prospects:The 2002 Revision, New York, United Nations.

United Nations Development Program UNDP (2000),Albanian Human Development Report 2000, Tirana,UNDP;

available online at:http://www.al.undp.org/download/pdf/nhdr2000.pdf,December 2003.

United Nations Development Program (2003), Bosniaand Herzegovina Human Development ReportMillennium Goals 2003, Bosnia and Herzegovina,UNDP;

available online at:http://www.undp.ba/publications/engleseka_final.pdf,December 2003.

United Nations High Commissioner for Refugees –UNCHR (2002), Statistical Yearbook 2001. Refugees,Asylum-seekers and Other Persons of Concern Trendsin Displacement, Protection and Solutions, Geneva,UNHCR; available online at: www.unhcr.ch, December2003.

United Nations High Commissioner for Refugees –UNCHR (2003), 2002 statistics on asylum-seekers,refugees and others of concern to UNHCR, Geneva,UNHCR; available online at: www.unhcr.ch, December2003.

World Bank (2003), World Development Indicators2003, World Bank.