Longer-Term Demographic Dynamics in South-East Europe: Convergent, Divergent and Delayed Development Paths
Author(s):Fassmann, HeinzMusil, ElisabethBauer, RamonMelegh, AttilaGruber, KathrinPublished in:Central and Eastern European Migration Review, Vol. 3, No. 2, 2014, pp. 150-172Views: 3900
This article offers an overview of the longer-term migratory and demographic developments in eight South-East European countries (Austria, Bulgaria, Italy, Hungary, Romania, Serbia, Slovenia and Slovakia). The main research question aims to analyse the different demographic historical developments and to examine whether convergent or divergent processes are dominant. Over the whole reference period, the population size in these eight South-East European countries (the SEEMIG region) grew from around 100 million people in 1950 to 122 million in 2011. This is surprising, as the public image of the region is linked to decline and backwardness and to being peripheral. However, major differences in the demographic developments of the countries can be observed. Some countries, including Austria, Italy and, with some fluctuations, Slovakia and Slovenia, experienced constant population growth during the entire reference period. All other countries were affected by a decrease in population, as was the case for Hungary in the early 1980s, Bulgaria at the beginning of the 1990s and Serbia and Romania since the start of the new millennium. The fertility trend shows a convergence while the mortality trends (including average life expectancy at birth) prove to be divergent. The net migration pattern seems to follow a migration cycle concept which postulates a general shift from emigration to immigration as a consequence of a declining natural increase on the one hand and a growing demand for new labour on the other. Some countries show trends that do not yet follow this pattern, which might indicate that additional factors and interpretative models should also be taken into account. The long-term distribution of growth and decline in the region is quite diverse and underlines the need for differentiation and specific explanations.
This paper offers an overview of the longer-term migratory and demographic developments in eight countries mainly from the Northern and North-Western parts of the South-East European region, in this study named as ‘SEEMIG region (or area).’ In the early 1950s, countries in this region of analysis had quite similar economic and socio-demographic patterns, especially, for example, in their GDP development or migratory trends. Over the reference period of the last 60 years, the Western and Eastern parts of the region developed quite differently, and the area developed into an inter-dependent and heterogeneous region. The main research question looks at these different demographic and migratory developments and investigates how legacies affect current demographic, economic and political trends and whether convergent or divergent processes are dominant in the region.
The article brings together the main findings of the country-level historical analyses of the eight countries (Austria, Bulgaria, Italy, Hungary, Romania, Serbia, Slovenia and Slovakia) of the SEEMIG project partnership – henceforth the SEEMIG countries. The analysis is based on longer-term national statistics on migration and other macro-statistical time series provided by the project partners as well as by the international databases of Eurostat, the United Nations, the World Bank and the Maddison project. An essential source of information was also provided by two recent publications carried out within the project partnership (Böröcz 2009; Melegh 2012), from which this paper draws conceptually as well as empirically. The results – which build on a synthesis analysis elaborated by Fassmann, Musil and Gruber (2014) – were additionally subject to thorough discussions within the SEEMIG partnership.
The conceptual framework1 of this historical analysis is based on several general and more theoretical observations. As such, the empirical work cannot be related to only one or two specific concepts. The concept of the so-called first ‘Demographic Transition’ (Landry 1934; Notestein 1945; Thompson 1929; and see Melegh 2006: 60–64, 71–76 for the history of the idea) is an important theoretical approach for the analysis as is the idea of ‘longue durée’ presented by Braudel (1969). In migration studies, the ‘Migration System Theory’ (Zlotnik 1992) and the historical-structural approach represented by Böröcz and Portes (1989) also influenced our analysis. For an overview on migration theories, see Massey (1999).
Apart from these general observations, two main conceptual approaches are employed. For the long-term analysis of migratory trends, we apply the concept of ‘migration transitions’ (de Haas 2010; Skeldon 2012; Zelinsky 1971) to analyse the relation between migration and broader development processes through space and time. As suggested by Skeldon (2012: 154), transition theory – while it has been disregarded in recent work on migration theory (Brettell, Hollifield 2000; Portes, DeWind 2007) – can be employed as a transitional framework, which allows migration systems to be linked to wider socio-economic change.
The transition model employed within our analysis is the ‘Model of the Migration Cycle’ (Fassmann, Reeger 2012). Based on evidence from several European countries, the concept is based on the idea that a society and the legal system of a country adapt to a new situation and develop a mechanism to handle new or evolving migratory circumstances (Fassmann, Reeger 2012: 67). This policy learning process, which becomes necessary when new demographic and economic conditions arise, is referred to as a migration cycle. According to labour migration theories, such as the push-and-pull model or migration systems theory, the main drivers stem from demographic, economic and sectorial development and the structuring of the labour market. The model thus describes the empirical observation that countries change, for example, from an emigration to an immigration country, if demographic reproduction is not guaranteed.
The article is organised as follows. In the first section, the main socio-economic, demographic and migratory dynamics are examined in a longer-term perspective reaching back to the 1950s. To differentiate, special focus is put on three major time periods that have been identified as particularly important for developments in the region. In the following section, the long-term development of fertility, mortality, net migration and population change is highlighted. The final section draws conclusions from the historical analysis and provides a brief reflection on the conceptual considerations.
Historical turning points and demographic periods – a literature review
Since 1950, the South-East European region has been characterised by several historical turning-points that affected countries within the region to different extents. As mentioned in the introduction, it is necessary to emphasise the historically given heterogeneity of the region of analysis. Whereas all other parts of Europe had more or less stable borders and political and economic systems, the Eastern and South-Eastern parts of Europe were affected by wars, post-war transformations and recovery, the construction and the breakdown of socialist regimes and its multinational socio-economic spaces, the dissolution of federal countries and changes in political regimes from state socialism to democratic capitalism. Recently, the countries also experienced accession or pre-accession arrangements to the greater economic and political space of the European Union. Most of the societal changes in South-East Europe are rooted in key long-term developments that exhibit time lags, intensifications and phases of slow-down as well as convergences with development in Western Europe. Nevertheless, although there are arguments which posit that most historical analyses emphasise political changes and patterns too strongly (Melegh 2012), especially when countries such as the SEEMIG countries, that represent varying political systems, are included in an analysis of long-term change, the importance of combined political and economic changes is indisputable. Three main historic turning points are highlighted in this regard:
- The installation of state socialism in several South-East European countries and the closing of the Iron Curtain in the late 1940s. This political reorientation had severe consequences, as it led to a new political, social and economic order in the countries concerned, including the centralised coordination of the economy and relatively closed borders vis-à-vis migration. In this regard it is important to stress, however, that there were different forms of socialism and therefore diverse impacts on the respective economies and societies.
- The breakdown of socialism and the fall of the Iron Curtain in 1989/1990 due to restructuring and the new cycle of the world economy and related internal development. The change in the political regime from socialism to democracy and the market transition in Bulgaria, Hungary, Romania and Czechoslovakia not only led to the shaping of market-like economies and dramatic changes in the labour market, but also strongly influenced the overall socio-demographic development of these countries by increasing unemployment and, simultaneously, creating possibilities for the free movement of the population, which altered the number and direction of migration flows. For example, in Bulgaria and Romania a rapid increase in the number of emigrants was noted during this period. As a further consequence of the breakdown of state socialism, new independent countries emerged, and international borders changed in Slovenia, Serbia and Slovakia which, again, led to increased migration.
- The step by step accession to the European Union (‘EU enlargement’) was a further turning-point that shaped socio-economic dynamics in the region. Integration into the second-largest economic area of the world brought an overall change of regimes in the countries, making it necessary to modernise their economies and implement imposed stability measures. With Italy being a founding member and Austria joining in 1995, the accession process in the other countries, evolved in the current millennium. In several SEEMIG countries – Austria, Italy, Slovenia and Slovakia – accession also meant the abandonment of independent national monetary policies. Accession further entailed the free movement of persons within the EU, step-by-step inclusion into the free movement area of Schengen, and harmonisation in various policy areas, including migration and asylum.
Politically divided South-East Europe, 1950–1989
In the countries of the so-called former ‘Eastern Bloc’ (Bulgaria, Hungary, Romania and Slovakia) and the Former Yugoslavia (Serbia and Slovenia),2 the installation of state socialism at the end of the 1940s induced major shifts in political, social and economic orders and paved the way for huge transformations. However, the developments in this region were much more differentiated than the term ‘Eastern Bloc’ would imply. While Bulgaria, Hungary, Romania and Slovakia, as members of the Council for Mutual Economic Assistance (COMECON), were oriented toward the Soviet Union politically and militarily, the Former Yugoslavia (and with it Slovenia and Serbia) started the ‘Non-Alignment Movement’ and held a special position within the Eastern European communist bloc. Austria and Italy moved towards Western Europe and experienced an economic boom labelled as an ‘economic miracle.’
The first decades after the Second World War were characterised by high fertility rates and population growth. This making up for ‘lost war years’ took place both in socialist countries of the Eastern Bloc (Bulgaria, Hungary, Romania and Slovakia) and in Austria and Italy. The high fertility rates in some countries were also partially supported by open and sometimes coercive pro-natalist policies. This included restrictions on or the prohibition of abortion as well as the installation of pro-natalist financial and social incentives in specific countries (e.g. family allowance, child-care benefits, a ‘tax on childlessness’ and an anti-abortion campaign in 1952–1953 in Hungary and strict anti-abortion measures in Romania after 1967). Nevertheless, these measures did not succeed unrestrictedly.
Where mortality and health patterns were concerned, after the high degree of convergence experienced with the generalised decline in infectious mortality across the industrialised world, the mid-1960s marked the start of a new divergence in life-expectancy changes (Meslé 2004a: 22). While, on the one hand, Western countries made rapid progress in the late 1960s after the more-or-less significant slowdown, the ground rapidly made up by Eastern European countries, on the other hand, was followed by a long period of stagnation or even deterioration. As such, Austria and Italy have moved in the longer term to a West-European model, with the exception of a few causes of death, which can mainly be explained by similar social developments and underlying social mechanisms (Mackenbach 2006; Meslé 2004a). In the other SEEMIG countries, a mortality and health crisis lasted until the end of the 1980s.
As regards migration trends during this period, Bulgaria, Hungary, Romania and Slovakia became relatively closed countries, though to different extents; Yugoslavia, meanwhile, officially promoted the temporary emigration of ‘guestworkers’ to Western countries. Restrictive border-control regimes came into force, meaning that cross-border migration was controlled in both directions and mostly repressed. Entering or leaving the country (even for tourism) was subject to special permits; possessing a passport and travelling abroad were limited rights and a privilege administered by and at the discretion of the authorities, for example in cases of family reunification.3 Those who left illegally or did not return home from abroad were sanctioned by being deprived of their citizenship, by the confiscation of their property or by imprisonment for illegal border-crossing.
In the context of these migration restrictions, unauthorised emigration became increasingly common during the last decade of socialism. These restrictions certainly played a major role in migratory changes, but it is important to note that other economic and historical factors of international migration were also influential. Major outflows occurred in the context of revolutions, such as that of 1956 and 1957 when some 176 000 people left Hungary, or in 1968, when 162 000 people left Czechoslovakia after the Prague Spring. These ‘refugees’ – although in many cases also economic migrants – were granted asylum by Western European countries, including Austria, which was the first safe country during the Cold War and thus a country of both destination and transit for refugees from Eastern Europe.
A further characteristic policy pursued in socialist countries was that of selective organised emigration. While the emigration of their own nationals was obstructed, it was common practice to organise and implement the emigration of ‘certain groups,’ such as certain ethnic minorities and political critics of the regime. As such, the mass migration of Jewish and German communities (e.g. Transylvanian Saxons and Banat Swabians) was promoted in Romania and organised and financially supported by Israeli and West German authorities. Likewise, in Bulgaria, migration processes were arranged by bilateral agreements between the Bulgarian and Turkish governments and expressed mainly through the emigration of persons of Turkish origin.4 As in the Eastern Bloc countries, controlled and ethnically selective emigration, based on the Balkan Pact, was also an essential feature in socialist Yugoslavia. As such, the controlled emigration of the majority of ethnic Germans in the early 1950s, followed by that of ethnic Turks and other Islamic ethnic groups, was organised from Serbia.5
Temporary labour migration schemes were a further feature of migration processes of this time which involved SEEMIG countries both as sending and as destination countries and which have formed migration processes ever since. Within the COMECON, a circular form of labour migration of professionals (e.g. engineers) and skilled workers (e.g. miners, pipeline constructors) existed in some member-states (especially Bulgaria, Hungary, Romania and Czechoslovakia). These flows were regulated by the states involved, and were intended to encourage temporary migration, not dissimilar from the labour schemes promoted by West Germany towards Hungary, Poland or Yugoslavia. At the same time, following intensive economic development in Western European countries that resulted in increased demand for an industrial labour force, so-called ‘guestworker’ schemes – the temporary employment of foreign workers – were launched, enabling a real expansion of international labour migration. Austria6 followed the example of Western neighbouring countries and started with the targeted recruitment of foreign workers from Mediterranean countries (especially Turkey and Yugoslavia) in the 1960s. The liberalisation of emigration regimes, the need for migrant workers in the West German economy and the agency of each individual triggered guestworker emigration from Slovenia to Western countries, especially West Germany. In Serbia,7 after restrictions on leaving the country were abolished in the mid-1960s, temporary labour migration abroad was completely liberalised. Temporary labour migrants were also recruited from Italy as low-skilled workers in the industrial sector in Northern European countries, again particularly West Germany.
Initially, the prevailing idea was that such migration would be temporary, but it soon became clear that many migrants continued to stay in their destination countries. Despite the attempts of both the destination countries – which aimed to reduce the number of foreign workers during the oil crisis of 1973 through the curtailing of recruitment – and the efforts of migrants’ countries of origin such as Serbia, to help returnees to start their own businesses, migrants remained abroad. In contrast to political intentions, these measures also partly led to a consolidation of the settlement of migrant workers. In Austria, fearing that they would lose their jobs and residence status following more restrictive measures, many migrants, who had been circulating between their country of origin and Austria, postponed their return. In parallel, family reunification compensated for the number of people who returned to their countries of origin.
With international migration restricted and industrialisation strongly pursued in many countries, large-scale internal migration flows were a further essential feature of the period, as the process of industrialisation induced intense concentration processes and a spatial redistribution of the population. The urbanisation process and the concentration of migrants were characteristic features during this time in Czechoslovakia, where internal migration between the later Czech and Slovak Republics was shaped by economic and industrial policies. These migration dynamics were significant not only in terms of their mutual influence on population development, but also in terms of the socio-economic development of both republics.8
Internal migration was also a characteristic feature in socialist Yugoslavia. In the 1970s in particular, when Western European countries that had been the primary countries of destination started limiting immigration as a response to the oil crisis, more-pronounced internal migration occurred within the former Yugoslav republics – especially Bosnia-Herzegovina, Serbia and Croatia – particularly towards Slovenia. This migration of mostly young, male migrants was promted by several developments: an accelerated industrialisation from the mid-1950s, when several hundreds of mostly unskilled workers moved to towns; unemployment in several former Yugoslav republics and almost full employment in Slovenia. Most migrants, although first labelled as temporary, later stayed. In Italy, too, the labour demand generated by economic growth was, for a time, satisfied by internal migration, especially in the 1950s and early 1960s, when major internal migration flows from the south of the country (the Mezzogiorno) ensued to the industrial north. This migration accelerated the urbanisation process in all countries.
Overall the relationship of these changes to longer-term cyclical changes also provides ample evidence that the full reintegration of highly industrialised socialist economies into competitive capitalism led to an increase in emigration, as explained by historical-structural approaches.
The fall of the Iron Curtain
Starting in the late 1970s, globalisation began to promote new types of economic policy and establish a new division of international labour, which gradually led to sweeping changes. In the late 1980s, a period of radical political and social transformations arose in the South-East European region. Besides political changes, the breakdown of socialist systems and the subsequent democratic transitions launched various economic and social changes in the SEEMIG countries. While in Bulgaria, Hungary, and Romania this occurred without changes to the integrity of the state, the dissolution of socialist Yugoslavia led to the appearance of Slovenia and Serbia, and the Czech Republic and Slovakia were also separated. The transition towards market-oriented economies, privatisation and the crisis of traditional industrial branches due to the rising new international division of labour meant the end of official full employment and the appearance of unemployment as a new social phenomenon.9 Employment rates worsened due to a lack of employment opportunities, particularly for certain population groups such as those with little education, young people, those of peri-retirement age and marginalised ethnic groups such as the Roma population. Early retirement was made possible for many workers instead of dismissal, which raised the already high share of pensioners. The transformation was also characterised by a massive decline of industry and a significant setback in economic prosperity, which were reflected by a sharp decline in GDP values between 1989 and 1993 as well as a rise in inflation. These changes also led to the comparatively relative decrease in the economic well-being and weight of these countries (Böröcz 2009).
Another significant consequence of the fall of the socialist systems was an increase in social inequality. While, prior to 1989, the Eastern European countries had been characterised by a very low level of income differences, these inequalities considerably increased after the collapse of the state socialism regimes. In spite of this trend, however, it is possible to distinguish between countries with a relatively low and those with relatively high level of income inequality. Inequality was found to be the highest in Hungary and Slovakia. For example, the increase in Hungarian income inequality during this period was found to exceed the increase experienced by both the United Kingdom over the 1981 to 1986 period and Sweden from 1988 to 1992 (Förster, Tóth 1997). After two decades of economic transition, Hungary, Slovakia and, above all, Slovenia were among the countries where the income inequality was relatively low. At the opposite end, Romania was characterised by quite high income differences.10
A further crucial consequence of the fall of the Iron Curtain was the loss of political control over migration. The national borders became permeable and the way to a free unfolding of migratory processes was opened. The end of the 1980s thus represents a turning point in migration in these countries. Unrestricted travelling from 1988 onwards and the permeability of the borders after the transition opened a new chapter in migration. As Melegh (2012) stresses, the fact that the economies of these countries were included in the space of global capitalism and that they were falling behind economically compared to global development had a significant impact on the migratory links of the countries within the region. After the fall of the Iron Curtain, the economic effects (de-industrialisation, unemployment and increasing wage differentials) were strong drivers within the migration system. However, with the exception of forms of temporary employment such as seasonal work programmes, labour migration to European Union member-states (many of which had a history stretching back to the state socialist period) was highly restricted in terms of visas, residency and employment. This had important consequences on the direction and the channelling of migration as ethnic and refugee migration. In Bulgaria, emigration largely affected the population development in a negative manner. In the period from 1985 to 1992, the population of the country decreased by 461 000 people. This decrease was almost entirely a consequence of the negative migration balance to Turkey, which received large numbers of people of Turkish origin from Bulgaria. According to estimates by Horváth and Kiss (2013: 17) based on mirror statistics and earlier studies (Gödri 2004; Szőke 1992), a total of 100 000 Hungarians left the Romanian region of Transylvania between 1988 and 1992, the majority of them going to Hungary. Following their analysis, Hungarians were also highly overrepresented among the emigrants between 1992 and 2002. Emigration from Hungary had a sharp peak immediately after the collapse of socialism, partly due to non-Hungarian citizens who had arrived in the country in previous years. This process was largely connected to transit migration from Romania and other countries to Western Europe. In this process, economic, political and ethnic factors have promoted out-migration in a complex and interdependent way (Melegh 2012).
Major outflows of refugees also arose from Romania in the last years of the repressive Ceauşescu regime and subsequent years. Hungary was the main destination for ethnic Hungarians; however, a part of them also migrated to Western Europe (including Austria, Germany and Sweden). In order to curb the flow of Romanian asylum-seekers, many Western European states amended their asylum laws by qualifying Romania as a ‘safe country of origin.’
In the same period, Hungary transformed from a country of emigration into a destination country of international migration.11 Immigrants came mostly from neighbouring countries, especially Romania, and also from Ukraine and Yugoslavia and its successor states. Between 1988 and 2007, approximately 200 000 foreign citizens received settlement permits (long-term or open-ended residence) (Póczik, Fehér, Dunavölgyi, Jagusztin, Windt 2008). In this way, a new institutional system of migration was established and legal frameworks, mostly of an administrative and law enforcement nature, were accordingly laid down.
Migration processes in the successor states of the Former Yugoslavia and their neighbouring countries were largely formed by the conflicts in the region. The wars at the beginning of the 1990s set off major flows of migration – over 5 million people are estimated to have been forced to leave their place of residence in the 1990s. While many returned later, a major share permanently changed their place of residence. Slovenia, Serbia and, partly, Hungary were among the main destinations for these forcibly displaced persons.
Emigration from the Former Yugoslavia was a further characteristic of the period. While emigration from Slovenia was relatively low, intensified emigration from Serbia during the 1990s could be observed. According to the 2002 census, 415 000 Serbian citizens (5.3 per cent of the total population) were registered as working or staying abroad – an increase of over 50 per cent in relation to the previous 1991 census – making the emigration of Serbian citizens in the late 1990s and the 2000s the most intense since the second half of the 1960s.
The new millennium opened up the perspective of integration into the European Union for most SEEMIG countries. Austria joined the EU in 1995, Hungary, Slovakia and Slovenia nine years later in 2004; Bulgaria and Romania followed in 2007. Serbia acquired official candidate status in the year 2012. Economic reforms initiated in the pre-accession and accession periods were directed towards a neo-liberal order: fiscal policies became stricter to conform to the EU accession criteria, privatisation processes were accelerated and the countries became more open to foreign actors. Consequently, the amount of foreign direct investment grew considerably. The SEEMIG countries in Eastern Europe experienced step-by-step integration into the common market and a convergence of the political and economic systems, which can be addressed as ‘Europeanisation.’ Distinctions have, however, to be made between the character of the transformation process of countries joining in 1995, including Austria, and the 2004/2007 EU enlargements. Since, overall, the countries involved provided socio-economic requirements which were even higher than the so-called EU standards, the accession evolved in a smoother manner (Böröcz 2000; Melegh 2006).
During the first years of the new millennium, the economies in the eastern SEEMIG countries (Bulgaria, Hungary, Romania, Slovakia and Slovenia) developed dynamically, with annual growth rates in GDP per capita that were, in certain years, five to six times higher than the EU27 average. The reasons given for these developments in the country reports included large-scale privatisation, foreign direct investment and reforms of public spending (e.g. austerity packages in Hungary). Although it is still being debated, the period from 2001 to 2011 is also perceived to be the decade which brought about the end of societal, institutional and economic transition. Nonetheless and despite these dynamics, the countries concerned could not reach their economic levels of the 1970s. As regards GDP per capita values compared to the world average, the relative position, for instance, of Romania in 2011 was still that of the 1950s. Finally, at the end of the decade in 2009, the global economic crisis caused a decrease of the GDP per capita, bringing about serious consequences in the whole region. The SEEMIG countries experienced a period of outright recession: GDP per capita declined by between 3.1 per cent (Serbia) and 8.8 per cent (Slovenia). In the last three years, after a period of stagnation, annual economic growth rates have increased again, but stayed at a lower level. One of the consequences of the economic crisis is the increasing unemployment and very low labour-force participation rates – a serious problem in nearly all SEEMIG countries.
The increased European integration also brought the free movement of persons within the area of the European Union: all EU citizens have the right to enter another EU member-state without an entry or exit visa for three months. Upon fulfilment of specific requirements,12 EU citizens may also remain beyond this period without a residence permit and may acquire the right to permanent residence in the host member-state after a five-year period of uninterrupted legal residence. The right to free movement, however, for countries joining in 2004 and 2007, was not accompanied by an immediate right to enter the labour market without any restrictions. Austria and Germany asked for a transitional phase of up to seven years, when restrictions on the freedom to enter the domestic labour market were abolished. In 2014, the last restrictions for other SEEMIG countries expired.
Furthermore, EU accession brought incremental inclusion into the territory of free movement of the Schengen Area. This possibility of free movement led to the creation of new labour markets, thus reinforcing inner-European mobility. A specific aspect of this includes the increase of cross-border commuters. Finally, former or continuing regions of emigration, especially the Ukraine and Russia, themselves became destinations for migrants outside the EU as well as a place of entry and a transit region for migrants and those from third countries in search of international protection.
Besides the legal changes described above, and in line with the proposed migration-cycle interpretation, the rise in the demand for immigrant labour in some core economies of the EU also played a role in migratory changes related to EU integration. Over and above internal adaptation, these modifications are also related to the changing position of the EU and some member-state economies, due to the intensification of global competition, as migrant labour provides lower wages.
Figure 1. Migration flows in Europe in 2005–2010
Source: Abel and Sander (2014); Visualisation: Sander, Bauer and Frank.
Due to the further integration of the SEEMIG region into the common area of the European migration space, it is not surprising that migration flows from and to the SEEMIG region are concentrated within Europe. The figure above shows SEEMIG flows within Europe between 2005 and 2010 and reveals that they are connected mostly to EU15 countries (i.e. EU member-states before 2004) as well as to other European countries, but not less to EU13 countries (i.e. new EU member-states since 2004).
Net flows in Austria were positive and relatively constant between 1990 and 2010. Only during the late 1990s were net gains lower due to both fewer in-flows and more out-flows. While the level of out-flows from Austria remained relatively stable after 1990, the origins of in-flows shifted from European non-EU countries to countries from the rest of the world. The in-flows from other EU countries were relatively constant but also shifted from EU13 to EU15 countries (mainly Germany). In Italy, net migration gains were moderate during the 1990s, but strongly increased after 2000. These changes in net flows can be attributed to a general drop in out-flows (especially to EU15 countries) and, first and foremost, to a strong influx from non-European countries.
The trend in net migration flows in the two Former Yugoslavian Republics in the SEEMIG region is not comparable by intensity or direction. Slovenia’s net flows were clearly positive between 1990 and 2010, while Serbia’s balance was positive during the early 1990s but became negative thereafter. Among the former COMECON countries in the SEEMIG region, it is necessary to distinguish between Hungary and Slovakia, with positive net migration flows on the one side, and Romania and Bulgaria – both clearly countries of emigration – on the other. Hungary’s net flows were almost constant during the four five-year periods between 1990 and 2010, but are actually a result of decreases in both out- and in-flows. The origin of in-flows also shifted from other EU13 countries to non-European countries. In Slovakia, the focus of both in- and out-flows clearly shifted from other EU13 countries (during the 1990s) to EU15 countries (since 2000), while the volume of out-flows steadily declined. Romania and Bulgaria experienced strong out-flows after 1990. Although the volume of out-migration has decreased since then – mainly due to fewer out-flows towards and more in-flows from non-European countries – the balance has remained clearly negative. In both countries, most out-flows after 1990 were directed to EU15 countries. Since 2000, Romania and Bulgaria have also received some in-flow from non-EU European countries and from the rest of the world, but rarely any from other EU countries.
Demographic developments in retrospect: convergence or divergence
After describing the overall demographic and migratory developments differentiated by time-period, we now concentrate on specific issues and variables such as convergence and divergence trends within the region. Measuring convergence presents a number of complexities, because several definitions exist that correspond to different concepts (Monfort 2008). For this analysis we restrict ourselves to assessing the dispersion of demographic developments at the national level. We therefore look at the variance (i.e. the statistical average of the amount of dispersion) as a measure of dispersal across the eight investigated countries, and how the values have changed over time. Convergence is hereby understood as a reduction in the coefficient of variance among regions, while divergence refers to a development with increasing variance. The change of population stock, fertility, mortality and net migration, as well as migration flows, are at the core of the analysis. While the concept of convergence lies at the heart of demographic transition theory (Wilson 2001: 155), this article does not suggest that the SEEMIG countries demographically converge in a nomothetic sense. Instead, it presents an idiographic analysis of demographic processes in the region.
Convergence fertility decline
The decline of the growth rate (see Figure 2) is mainly caused by natural population decrease. At the beginning of the reporting period, the total fertility rate in the region was approximately 2.7 children per woman. It remained at around 2.5 until the mid-1970s and experienced – more than ten years after the birth decline in Western Europe – a moderate decrease to 1.8 children per woman. The fertility rate then sharply declined and has remained below 1.5 since then. Only the Serbian TFR (1.6 in 2010) remained above the SEEMIG average. At the beginning of the reporting period, the variation coefficient of the total fertility rate was around 17 – this means that the standard deviation was about 17 per cent of the mean. The variation coefficient declined to 9 in the first half of the 1970s and then increased to 19 in the second half of the 1980s. Hungary, Serbia, Slovakia and, in particular, Romania were able to successfully slow down their fertility decline, but others were not. However, after the fall of the Iron Curtain, the variance declined significantly and the variation coefficient decreased to 8 in 2010.
The so-called ‘Socialist Greenhouse Theory’ offered by Sobotka (2002) explains this time delay in European post-communist countries and describes the general social, political and institutional environment of socialist states aiming to stabilise reproduction rates. The fertility differentials in the region increased due to restrictive birth-control regimes in some countries and wide-scale pro-natalist and social support in other state socialist countries, which pushed fertility higher there than in, for example, Italy and Austria. Whether or not these changes indicate a second demographic transition or demographic crisis is subject to current academic debates (Rychtarikova 2001). However, the high unemployment rate and the relative deprivation of larger households clearly signal that there will be no return to a high fertility level. The convergence tendency of the total fertility rate and the resulting narrowing of the variance of the rate emphasise the generality of the trend.
Figure 2. Total fertility rates in 1950–2010
Source: UN World Population Prospects: The 2010 Revision, own illustration.
Time-lagging net migration
Another factor that influenced the slow-down of the population growth rate is migration. The country reports and data provided by the United Nations13 allow for a long-term analysis of migration trends in the reporting countries (see Figure 3). Major changes in the migratory trends emerged within the region. At the beginning of the reporting period, most of the SEEMIG countries showed a net migration below zero, meaning that emigration in that time-period exceeded immigration. Beginning in the 1960s, however, the region lost this homogeneity and, in the following decades, the SEEMIG countries were increasingly characterised by a diversification of their net migration rates. While some regions became immigration areas, others became or remained areas of emigration (Melegh 2012). This is also reflected in the variation coefficients – they are much higher and more fluctuating compared to the total fertility or the life expectancy.
Like other Western European countries, Austria and, later, Italy and Slovenia experienced favourable economic development and turned into de facto immigration countries. The active recruitment of foreign workers prevailed as labour-market measures in order to satisfy labour-market demands. Romania, Bulgaria and Serbia, in particular, became emigration countries, whereas countries like Hungary and Slovakia – according to the officially registered net migration rates – turned into ‘emerging immigration countries.’ There, a positive migration balance was reached in the 1980s and in countries like Hungary it could probably be easily reversed by recent trends.
Figure 3. Net migration rates in SEEMIG countries (per 1 000) in 1950–201014
Source: UN World Population Prospects: The 2010 Revision, own illustration.
Based on the findings of the country reports and net migration trends from 1950–2010 and applying the migration-transitions approach to the SEEMIG countries, three types of national migration status can be identified: Type 1 = relatively ‘old immigration countries,’ Type 2 = ‘emerging immigration countries,’ and Type 3 = ‘emigration countries’ (see Table 1). These types are characterised by specific migration patterns and policy attitudes.
- Type 1 ‘old immigration countries’ are located at the end of an adaptation cycle regarding immigration. These countries represent the overall European pattern of development: they showed a negative migration rate in the 1950s, but their migration rates became positive parallel to the process observed on the entire continent. As the ‘tipping point’ had already been reached several decades earlier, they have learned to treat immigration politically, instrumentally and in public discourse. In the SEEMIG region, Austria and Slovenia belong to this group, next to other countries such as Belgium, Denmark, Germany, the Netherlands and Sweden.
- Type 2 ‘emerging immigration countries.’ In these countries a positive migration balance has only recently been reached. The topic of immigration is very contentious in public discourse and the adaptation process is at a very early stage. While Italy joined this group in the 1970s, Hungary and Slovakia have also belonged to this group since the early and late 1990s, alongside countries in Southern Europe such as Greece, Spain and Portugal, as well as Ireland and Finland which, especially before the global economic and financial crisis, were recently confronted with significant immigration.
- The so-called Type 3 ‘emigration countries’ are still characterised by major emigration flows. However, it is likely that these countries could also become countries of immigration in the future. Countries of Eastern Europe belong to this stage of constant and negative migration balance, amongst which Bulgaria, Romania and Poland. During the times of socialism and mobility constraints, immigration was not a major topic, not the least because the birth decline set in at a later stage.
Table 1. Typology of SEEMIG countries by migration status, in 2010
a With the exception of major refugee inflows during wars in Yugoslavia in the early 1990s.
Source: Fassmann and Reeger (2008) modified and amended.
It is important to note that, even on a regional level, these patterns of change can be related to and reinforce each other in the form of unequal exchange between the various sub-regions.
Divergent trends in life expectancy
The third main demographic factor that must be presented and analysed is the development of mortality. Two different observations must be underlined: the first is the increasing life expectancy of both males and females in the entire SEEMIG region (see Figures 4, 5 and 6). Over the whole period, the life expectancy at birth grew by around eight additional years or 1.4 years each decade. Nevertheless, while for the past 60 years all Western European countries, including Austria and Italy, have shown increases in life expectancy, Eastern European countries have had a different and altogether more negative experience. After a period of improvement due to better health care between the late 1980s and the late 1990s, these long-term negative trends in Eastern European countries – for example in Slovakia and Hungary – were reversed for certain male age groups, which led to an overall stagnation of mortality. The gain of additional years of life is higher in the period after the fall of the Iron Curtain than before. In contrast to increasing unemployment or people living below the poverty line, the growth in life expectancy is a positive consequence of the transformation process.
Figure 4. Life expectancy at birth, males and females (years), in 1950–2010
Source: UN World Population Prospects: The 2010 Revision, own illustration.
However, differentiation is again necessary, because a growing divergence in the increase of life expectancy is apparent (Meslé 2004a). The variation coefficients reached their relative minimum in the early 1970s at only 1.6. The standard deviation as a percentage of the mean increased to 4.5 in 2010, which is low compared to those of fertility or migration, but high within the whole reference period. People in countries like Austria and Italy are gaining more years of life than – for example – those in Bulgaria, Romania and Serbia. While, in 1950, at the beginning of the reporting period, the difference in life expectancy between Austria and Romania was around five years, by the end of the period, the difference had grown to around seven years. For the whole period, the population in Austria experienced an increase of 14 years, while that of Romania gained only 12 years. Differentiation by gender is also necessary. While men in general show a lower life expectancy than women, the mortality crisis of men in Eastern Europe has been subject to extensive academic research. While female life expectancy increased on average by 13.9 years in the SEEMIG region, the life expectancy of men has grown by only 11.2 years since 1950. The gender gap in life expectancy at birth increased on average from 3.8 years in 1950 to 6.2 years in 2010.
In general, the increase in lifespan is due more to a decrease in infant mortality than to additional years of life in old age. In the 1950s and 1960s, in particular, infant mortality in all countries of the SEEMIG region was reduced significantly through investment in medical services. Convergence and a decline of the infant mortality rate below 10 per 1 000 can be observed for the whole period and without historical breaks, although significant differences are still prevalent at the NUT3 level. Furthermore, favourable trends that occurred in Western Europe as long ago as the early 1970s have spread to Eastern European countries. While it is difficult to assess the main causes of such a reversal, the progress may result from a combination of several factors, including changes in diet, the growth of systematic prevention and screening and the spread of new forms of treatment and cardiac surgery (Meslé 2004b).
Population stock and population change
Over the reference period, the population size in the SEEMIG region grew from around 100 million people in 1950 to 122 million in 2011. This is surprising because the public image of that region is linked to decline, backwardness and a peripheral status. However, major differences in the demographic developments of countries can be observed.
Some countries experienced constant population growth during the entire reference period, including Austria, Italy and, with some fluctuations, Slovakia and Slovenia. All other countries were affected by population decrease. This was the case for Hungary and Bulgaria from the beginning of the 1990s and has also been the case for Serbia and Romania since the beginning of the new millennium. The long-term distribution of growth and decline is quite diverse and underlines the need for differentiation and specific explanations.
Table 2. Population stock on 1 January in SEEMIG countries, 1951–2011 (in thousands)
Notes: Figures for 1951 for Austria and Italy from 1952; for Slovakia from 1950; for Serbia and Slovenia from 1953. Values for Serbia for 2001 from 2002; for Romania for 1951, 1961, 1971, 1981 from UN World Population Prospects: The 2012 Revision, values for mid-year.
Source: National Statistical Offices, SEEMIG WP3 Country Reports.
As shown in Table 2, the SEEMIG region as a whole has been one of growth; however, two things are apparent. First, this growth in absolute terms was mainly due to immigration to countries within the region such as Austria and Italy, immigration which stemmed from other countries within the region. Secondly, the speed of growth is declining (see Figure 5). At the beginning of the reporting period, the annual growth rate exceeded 0.5 per cent each year. After a period of slow decline, the growth became negative after the fall of the Iron Curtain. A significant decline in the birth rates and increasing emigration flows were responsible for this development. Since the turn of the century, the growth rate has recovered and caught up to the zero line. However, once again, the large differences between countries must be mentioned. On the one hand are the steadily growing countries of Austria and Italy, which have been joined again in this trend in the last two decades by Slovenia; on the other hand are Bulgaria, Hungary, Romania and Serbia, which have experienced population decrease.
Figure 5. Average annual rate of population change (percentage) in 1950–2010
Source: UN World Population Prospects: The 2010 Revision, own illustration.
As another consequence of birth decline, of the emigration of the mostly younger population and of increasing life expectancy, a process of population ageing can be observed in the whole SEEMIG region (see Figure 6). The convergence tendency is to some extent parallel in all countries, excluding the case of Slovakia. This becomes apparent when looking more closely at the data: in the 1960s, Austria was at the forefront of the ageing process, over and above all other SEEMIG countries. With the return of high net migration in the 1990s and 2000s, the share of elderly people as a percentage of the working-age population decreased, but not because of the emigration of the older population or to increasing mortality. The working-age population itself increased in absolute numbers as a consequence of immigration and, therefore, the old-age–dependency ratio decreased. The opposite development can be observed in countries like Bulgaria. Due to the emigration of the working-age population, the share of the older population saw a relative over-proportional increase. While ageing is still one of the major demographic characteristics in the country (Vaňo et al. 2013), Slovakia nevertheless shows a different development – that of a stagnating age–dependency ratio of 17 per cent – and, as such, still holds the lowest share of elderly people within the working-age population of the region.
Figure 6. Age–dependency ratio, old (per cent of working-age population) in 1950–2012
Source: World Bank, World Development Indicators.
This dynamic historical analysis, carried out for the SEEMIG region for the period 1950 to 2013, has shown that these years were characterised, to differing degrees, by several historic turning-points in countries within the region and have contributed to a historically given heterogeneity: wars, post-war transformations and recovery, the breakdown of socialist regimes, the dissolution of countries and changes in political regimes from state socialism to capitalist democracy, have all affected the countries. Recently, (pre-)accession to the greater economic and political space of the European Union has marked regional developments. One of the main findings in this regard is that the influence of political systems can be observed and should not be neglected; however, varying trajectories of social and economic development were noted, even among state socialist countries. The different social histories of path dependencies, and global structural positions would appear to contribute to differences between regions, as they may have their roots in the historically accumulated structural differences (Chirot 1991).
General trends of convergence and divergence in the region can be traced as far back as the 1950s. Trends of convergence relate, for instance, to fertility; with some delay in time, a convergence tendency in the total fertility rate and the resultant narrowing of the variance of the rate both emphasise the general trend towards low fertility throughout the region. The total fertility rate declined sharply over time in all countries and has remained at a low level – below 1.5. The high unemployment rate and the relative deprivation of larger households clearly signal that there will be no return to a high fertility level. The SEEMIG region also reflects the overall European trend of population ageing. Throughout all SEEMIG countries, the proportion of the population of working age has been consistently decreasing (or stagnating) and reached a level of 66.5 per cent in 2011, while the share of persons of retirement age climbed to 17.7 per cent (EU27: 15.5 per cent). The changes in the age structure are irreversible and entail a range of challenges for public budgets in securing services of general interest, ensuring individual mobility through public transport, and maintaining pension and health-care systems. This also implies a societal and inter-generational potential for conflict as well as direct and indirect implications for the labour market.
One of the trends of divergence is noted in the area of mortality: while an increasing life expectancy can be observed throughout the region, a growing divergence between the countries is apparent. Divergences are also prevalent regarding economic developments: while the entire SEEMIG region is experiencing economic growth from a long-term perspective and has improved its position against the world average – with some countries producing three times more GDP than the world average and others only producing half – the main differences in the economic performance of countries within the region are predominant.
These differences might also have influenced the fact that net migration also showed different patterns; starting in the 1960s, the SEEMIG countries were progressively characterised by a diversification of net migration rates. While some countries became immigration countries, others became or remained emigration countries. Migration as such is a key driver of population development in the region. In particular, after 1989, the temporary emigration abroad started to be a key differentiating factor among the six former socialist countries. The recent overall population growth of the region can be almost exclusively traced back to immigration to countries in the region such as Austria and Italy, immigration which stemmed from other countries within the region. These connections also show intra-regional links between various developmental patterns.
The trends in international net migration are neither divergent nor convergent. They follow partially overlapping and partially time-lagging migration cycles with the changing dominance from emigration to immigration. Following the empirical analysis, Austria and Slovenia belong to the group characterised as ‘old immigration countries.’ Countries in this type, which is located at the end of an adaptation cycle, show the general European pattern of development: they had a negative migration rate in the 1950s which then became positive, corresponding with the overall observable change in Europe. As the ‘tipping-point’ was reached several decades earlier, they have learned to treat immigration politically, instrumentally and in public discourse (Fassmann 2009). According to the same typology, Italy, and partly also Hungary and Slovakia, were classified as ‘emerging immigration countries,’ representing countries situated in the transitional phase. In these countries a positive net migration was reached in the late 1980s and early 1990s. Since immigration is still a very contentious topic in public discourse, the adaptation process seems to be at a very early stage. Countries of the group ‘emigration countries,’ which include Bulgaria and Romania, are still marked by a constant and negative migration balance. Consequently, emigration and not immigration is the focus of public discourse. However, in the context of continued negative natural population development, emigration cannot represent an infinite process in these countries. These population developments rather suggest that these countries will also, in the long run, increasingly become destinations for immigration. Nonetheless, we must stress that, depending on future macro-structural changes, the previously differing historical trajectories and the varying modes of incorporation into the European and the global economy might maintain divergences in the longer run. What our analysis also shows is that, even in the regions with state socialism, various patterns of development existed. Some countries, such as Slovenia, were quite similar to non-socialist countries, such as Austria, while there were substantial difference between socialist countries like Romania and Hungary.
The presentation of the longer-term perspective has proven to be a challenge, as the aim of the analysis was to elaborate a comprehensive picture of long-term processes in a heterogeneous region in a period of somewhat dramatic structural changes. However, we follow Skeldon (2012: 154) in his conclusion that, While no single pathway through any migration or developmental transition exists, it nevertheless needs to be accepted that a retreat to total relativism is counterproductive. (…) a transitional framework, which allows migration systems to be linked to wider socio-economic change, provides a fertile environment in which to generate future theories of migration. Therefore it seems scientifically interesting to link migration with macro-structural changes to understand the different developmental tracks. As demographic and migratory trends proof to show very different patterns and pictures depending on the scale of observation there is also a need for the study on a sub-national level, particularly in this heterogeneous region. In this context, however, the particular poor situation of data quality, availability but also harmonisation or migration flows (Fassmann, Reeger, Sievers 2009, Lemaitre 2005, Poulain, Perrin, Singleton 2006) has to be highlighted again. To enable founded research in this direction but also evidence-based policy-making, better data not only on the national but also the sub-national and local levels are imperative.
1 For further information, see Fassmann and Musil (2013).
2 The Slovak Republic was part of Czechoslovakia until 1992; the Socialist Republic of Serbia and the Socialist Republic of Slovenia were part of the Socialist Federal Republic of Yugoslavia.
3 Politically favoured groups, such as Greek refugees in 1949 and Chilean refugees in 1973 in Hungary, who received settlement permits, were an exception to these rules.
4 An exception was the emigration registered in 1989 due to the obligatory change of the names of Bulgarian citizens of Turkish origin.
5 It is estimated that around 40 000 ethnic Germans, mainly from Vojvodina, emigrated to Austria and Germany; 250 000 ethnic Turks went to Turkey.
6 There was a need for foreign labour in the context of a booming economy and an inadequate work force because of the loss of men during the Second World War, post-war emigration and a decreasing female labour participation rate during the baby boom.
7 According to the results of censuses carried out in 1971, 1981 and 1991, the number of Serbian citizens working or staying abroad continually increased (from 204 000 to 269 000 and then to 274 000), which meant that approximately every thirtieth Serbian citizen lived abroad at that time (from 2.8 per cent in 1971 to 3.5 per cent in 1991).
8 During the existence of Czechoslovakia, 679 500 people emigrated from the Slovak to the Czech Republic and 440 000 people emigrated from the Czech to the Slovak Republic.
9 The process of privatisation and economic readjustment to world capitalism ranged from shock therapy in Estonia, at one extreme, to a very smooth and gradual transition in Slovenia, at the other. On this continuum, Hungary stood closer to the ‘shock therapy’ endpoint, contrary to other Visegrad countries that privatised their economy more gradually. In total, 1.5 million jobs were lost in Hungary and never recovered.
10 EU-SILC figures on the Gini coefficient of equivalised disposable income in 2013 were 35.4 in Bulgaria, 34 in Romania and 32.5 in Italy, above the EU28 average of 30.5. Slovakia (24.2), Slovenia (24.2) and Austria (27) had a comparatively low unequal distribution of income. With an income quintile share ratio of 6.6 in 2013, Bulgaria and Romania were also among the top three European Union member-states with the highest inequality of income distribution. Slovenia and Slovakia, however, both with a ratio of 3.6, were among the five countries with the lowest inequality of income distribution within the European Union.
11 At the same time, a constant flow of emigrants also existed, not registered by Hungarian statistics but clearly reflected in the relevant mirror statistics.
12 The conditions include, for example, the engagement in economic activity (on an employed or self-employed basis); the presence of sufficient resources and health insurance; the attendance at vocational training as a student and the presence of sufficient resources and health insurance; or being a family member of an EU citizen who falls into one of the above categories.
13http://esa.un.org/unpd/wpp/Excel-Data/migration.htm (accessed: 18 December 2014).
14 These net migration numbers are blurred and do not show trends regarding immigration and emigration separately. Furthermore, emigration data are often unreliable. For further information see Gárdos, Gödri (2014).
Abel G. J., Sander N. (2014). Quantifying Global International Migration Flows. Science 343(6178): 1520–1522.
Albani M., Callà R., Guarneri A., Piovesan S., Rattin C., Reynaud C. (2013). Dynamic Historical Analysis of Longer Term Migratory, Labour Market and Human Capital Processes in Italy. Country report developed within the project ‘SEEMIG Managing Migration and Its Effects – Transnational Actions Towards Evidence Based Strategies.’ Online: http://www.seemig.eu/downloads/outputs/SEEMIGHistoricalAnalysisItaly.pdf (accessed: 3 March 2014).
Böröcz J. (2000). The Fox and the Raven: The European Union and Hungary Renegotiate the margins of ‘Europe.’ Comparative Studies in Society and History 42(4): 847–875.
Böröcz J. (2009). The European Union and Global Social Change: A Critical Geopolitical-Economic Analysis. London: Routledge.
Böröcz J., Portes A. (1989). Contemporary Immigration: Theoretical Perspectives on its Determinants and Modes of Incorporation. International Migration Review 10(1): 606–630.
Braudel F. (1969). Histoire et sciences sociales. La longue durée, in: F. Braudel (ed.), Écrits sur l’histoire, pp. 41–83. Paris: Flammarion.
Brettell C. B., Hollifield J. F. (eds) (2000). Migration Theory: Talking Across Disciplines. New York: Analysis and London: Routledge.
Chirot D. (1991). The Crisis of Leninism and the Decline of the Left. Seattle: University of Washington Press.
Cukut Krilić S., Novak T., Jurišić B. (2013). Dynamic Historical Analysis of Longer Term Migratory, Labour Market and Human Capital Processes in Slovenia. Country report developed within the project ‘SEEMIG Managing Migration and Its Effects – Transnational Actions Towards Evidence Based Strategies.’ Online: http://www.seemig.eu/downloads/outputs/SEEMIGHistoricalAnalysisSlovenia.pdf (accessed: 3 March 2014)
de Haas, H. (2010). Migration Transitions: A Theoretical and Empirical Inquiry into the Developmental Drivers of International Migration. IMI Working Paper 24 (DEMIG Project Paper 1). Oxford: International Migration Institute.
Fassmann H. (2009). Von jungen und alten Einwanderungsländern: Die Geographie der europäischen Migration. Mitteilungen der Österreichischen Geographischen Gesellschaft 151: 9–32.
Fassmann H., Musil E., Gruber K. (2014). Dynamic Historical Analysis of Longer Term Migratory, Labour Market and Human Capital Processes in the SEEMIG Region. SEEMIG Working Paper 3. Budapest: Hungarian Demographic Research Institute.
Fassmann H., Musil E. (2013). Conceptual Framework for Modelling Longer Term Migratory, Labour Market, and Human Capital Processes. SEEMIG Working Paper 1. Vienna, University of Vienna. Online: http://www.seemig.eu/downloads/outputs/SEEMIGWorkingPapers1.pdf (accessed 12 December 2014).
Fassmann H., Musil E., Gruber K. (2013). Dynamic Historical Analysis of Longer Term Migratory, Labour Market and Human Capital Processes in Austria. Country report developed within the project ‘SEEMIG Managing Migration and Its Effects – Transnational Actions Towards Evidence Based Strategies.’ Online: http://www.seemig.eu/downloads/outputs/SEEMIGHistoricalAnalysisAustria.pdf (accessed: 3 March 2014)
Fassmann H., Reeger U. (2008). ‘Old’ Immigration Countries. Synthesis Report. IDEA Working Paper 3. Online: http://www.idea6fp.uw.edu.pl/pliki/WP3_Old_countries_synthesis.pdf (accessed: 3 March 2014).
Fassmann H., Reeger U. (2012). Old Immigration Countries in Europe. The Concept and Empirical Examples, in: M. Okólski (ed.), European Immigrations. Trends, Structures and Policy Implications, pp. 65–90. Amsterdam: Amsterdam University Press.
Fassmann, H., Reeger U., Sievers W. (eds) (2009). Statistics and Reality: Concepts and Measurements of Migration in Europe. Amsterdam: Amsterdam University Press.
Förster M., Tóth I. G. (1997). Poverty, Inequalitites and Social Policies in the Visegrad Countries. Economics of Transition 5(2): 505–510.
Gárdos É., Gödri I. (2014). Analysis of Existing Migratory Data Production Systems and Major Data Sources in Eight South-East-European Countries. SEEMIG Working Papers 2. Budapest: Hungarian Demographic Research Institute.
Gödri I. (2004). A magyarországra bevándorolt népesség jellemzői, különös tekintettel a Romániából bevándorlókra, in: T. Kiss (ed.) Népesedési folyamatok az ezredfordulón Erdélyben, pp. 126–147. Cluj-Napoca: Kriterion Könyvkiadó–RMDSZ Ügyvezető Elnökség.
Gödri I., Soltész B., Bodacz-Nagy B. (2013). Dynamic Historical Analysis of Longer Term Migratory, Labour Market and Human Capital Processes in Hungary. Country report developed within the project ‘SEEMIG Managing Migration and Its Effects – Transnational Actions Towards Evidence Based Strategies.’ Online: http://www.seemig.eu/downloads/outputs/SEEMIGDataSystemsCountryReportHungary.pdf (accessed: 3 March 2014).
Horváth I., Kiss T. (2013). Dynamic Historical Analysis of Longer Term Migratory, Labour Market and Human Capital Processes in Romania. Country report developed within the project ‘SEEMIG Managing Migration and Its Effects – Transnational Actions Towards Evidence Based Strategies.’ http://www.seemig.eu/downloads/outputs/SEEMIGHistoricalAnalysisRomania.pdf (accessed: 9 December 2014).
Landry A. (1934). La Révolution Démographique – Études et Essais sur les Problèmes de la Population. Paris: INED-Presses Universitaires de France.
Lemaitre G. (2005). The Comparability of International Migration Statistics Problems and Prospects.
OECD Statistics Brief. Paris: OECD. Online: http://www.oecd.org/migration/49215740.pdf (accessed: 10 October 2014).
Lukić V., Marinković I., Nikitović V., Ostojić I., Penem G., Predojević–Despić J., Rašević M., Stanković B., Šobot A., Zvezdanović J. (2013). Dynamic Historical Analysis of Longer Term Migratory, Labour Market and Human Capital Processes in Serbia. Country report developed within the project ‘SEEMIG Managing Migration and Its Effects – Transnational Actions Towards Evidence Based Strategies.’ Online: http://www.seemig.eu/downloads/outputs/SEEMIGHistoricalAnalysisSerbia.pdf (accessed: 3 March 2014).
Mackenbach J. P. (2006). Health Inequalities: Europe in Profile. Online: http://www.who.int/social_determinants/resources/european_inequalities.pdf (accessed 28 November 2014).
Massey D. S. (1999). Why Does Immigration Occur? A Theoretical Synthesis, in: C. Hirschman, P. Kasinitz, J. DeWind (eds), The Handbook of International Migration: The American Experience, pp. 34–52. New York: Russell Sage Foundation.
Melegh A. (2006). On the East-West Slope. Globalization, Nationalism, Racism and Discourses on Central and Eastern Europe. Budapest: Central European University Press.
Melegh A. (2012). Net Migration and Historical Development in Southeastern Europe since 1950. Hungarian Historical Review 1(3–4): 415–453.
Melegh A., Thornton A., Philipov D., Young–DeMarco L. (2013). Perceptions of Societal Developmental Hierarchies in Europe and Beyond: A Bulgarian Perspective. European Sociological Review 29(3): 603–615.
Meslé F. (2004a). Mortality in Central and Eastern Europe: Long-Term Trends and Recent Upturns. Demographic Research, Special Collection 2: 45–70.
Meslé F. (2004b). Écart d’espérance de vie entre les sexes: les causes du recul de l’avantage féminin. Revue d’épidémiologie et de santé publique 52(4): 333–352.
Monfort P. (2008). Convergence of EU Regions. Measures and Evolution. Working Paper 1, European Union Regional Policy. Online: http: ec.europa.eu/regional.../200801_convergence.pdf (accessed: 21 November 2014).
Notestein F. W. (1945). Population: The Long View, in: T. W. Schultz (ed.) Food for the World, pp. 36–57. Chicago: University of Chicago Press.
Pacheva A., Dimitrova D., Toteva D., Bakalova E., Yakimova E., Kostova M., Naydenova D. (2013). Dynamic Historical Analysis of Longer Term Migratory, Labour Market and Human Capital Processes in Bulgaria. Country report developed within the project ‘SEEMIG Managing Migration and Its Effects – Transnational Actions Towards Evidence Based Strategies.’ Online: http://www.seemig.eu/downloads/outputs/SEEMIGHistoricalAnalysisBulgaria.pdf (accessed: 3 March 2014).
Póczik S., Fehér L., Dunavölgyi S., Jagusztin T., Windt S. (2008). Nemzetközi migráció – nemzetközi kockázatok. Magyar Tudomány 2008/9: 1095–1107. Online: http://www.matud.iif.hu/08sze/07.html (accessed: 30 November 2014).
Portes A., DeWind J. (2007). Rethinking Migration: New Theoretical and Empirical Perspectives. New York: Berghahn Books.
Poulain M., Perrin N., Singleton A. (2006). THESIM – Towards Harmonised European Statistics on International Migration. Louvain-la-Neuve: UCL Presses Universitaires.
Rychtarikova J. (2001). Demográfiai átmenet zajlik Kelet-Európában? Is There a Second Demographic Transition in Eastern Europe? Regio 12(1): 111–139.
Skeldon S. (2012). Migration Transitions Revisited: Their Continued Relevance for the Development of Migration Theory. Population, Space and Place 18(2): 154–166.
Sobotka T. (2002). Ten Years of Rapid Fertility Changes in the European Post-Communist Countries. Evidence and Interpretation. Working Paper Series 02-1. Groningen: Population Research Centre.
Szőke L. (1992). Hungarian Perspectives on Emigration and Immigration in the New European Architecture. International Migration Review 26(2): 305–323.
Thompson W. S. (1929). Population. American Journal of Sociology 34(6): 959–975.
United Nations Population Division (2011). World Population Prospects: 2010 Revision. New York: United Nations, Department of Economic and Social Affairs.
United Nations Population Division (2013). World Population Prospects: 2012 Revision. New York: United Nations, Department of Economic and Social Affairs.
Vaňo B., Bleha B., Divinský B., Jurčová D., Katerinková M., Mészáros J., Pilinská V., Šprocha B., Bugár B. (2013). Dynamic Historical Analysis of Longer Term Migratory, Labour Market and Human Capital Processes in Slovakia. Country report developed within the project ‘SEEMIG Managing Migration and Its Effects – Transnational Actions Towards Evidence Based Strategies.’ Online: http://www.seemig.eu/downloads/outputs/SEEMIGHistoricalAnalysisSlovakia.pdf (accessed: 3 March 2014).
Wilson C. (2001). On the Scale of Global Demographic Convergence 1950–2000. Population and Development Review 27(1): 155–171.
World Bank. (various years) World Development Indicators. Online: http://data.worldbank.org/data-catalog/world-development-indicators (accessed: 3 March 2014).
Zelinsky W. (1971). The Hypothesis of the Mobility Transition. Geographical Review 61(2): 219–249.
Zlotnik H. (1992). Empirical Identification of International Migration Systems, in: M. M. Kritz, L. L. Lim, H. Zlotnik (eds), International Migration Systems: A Global Approach, pp. 19–40. Oxford: Clarendon Press.