Emigration From Post-Communist Central Europe After 1989 Interpreted Within the Aspirations/Capabilities Framework *

: In the period of post-communist transition, Central Europe witnessed complex and multifaceted mobility processes; permanent outmigration, of an ethnic or labour-related nature, coexisted with temporary, seasonal, or cross-border movements and an increasing in ﬂ ux of foreigners. To study these complex processes, we have chosen to apply a holistic and comprehensive approach, rather than limit conceptual considerations to one theory of migration determinants. We focus on eleven post-communist countries that joined the European Union (EU-11) and on the period extending from around 1989, covering the EU’s eastward enlargement, to the present. The aim of this study is twofold: ﬁ rst, we propose a general conceptual framework, based on the aspirations/capabilities approach, to present the main determinants of emigration from this part of the European continent. Second, in relation to each determinant, we formulate research questions postulated by selected theories of international migration and present the evidence, based on existing empirical studies, that addresses these questions. The paper contributes to the literature by providing a broad interpretation of post-transition mobility and pointing to commonly overlooked explanatory factors. We highlight the importance of economic factors that have enhanced and directed the outward migration from the EU-11 to selected EU member states and selected economic sectors; in particular, as regards capabilities, these factors include the lifting of labour market restrictions, high demand in the secondary sector of labour markets, and the roles of migration networks and the migration industry. Emphasis is also placed on aspirational factors, such as labour market failures and the substantial aspirational gap resulting from improvements in high educational attainment in the countries of origin. The aspirations/capabilities approach serves well as a general framework of migration determinants, but its explanatory power is enhanced by reference to other, more speci ﬁ c theories of migration. We show that a combination of the complementary approaches provides a more re ﬁ ned and in-depth picture of migration from the region.


Introduction
A historical study on international migration during the communist era in Poland was entitled "A country with no exit?",with a signifi cant question mark at the end of the sentence (Stola 2010).Indeed, for more than three decades -from the 1950s to the mid-1980s -international mobility was highly regulated and restricted not only in Poland but in all other communist countries of Central Europe (hereinafter CE), and only occasional waves of emigration for ethnic and/or political reasons and short-term commercial mobility behind the Iron Curtain justify the interrogative form of the book's title.The change of political regime at the turn of the decade from the 1980s to 1990s released the migration potential that had been building up for so many years, and international mobility increased exponentially throughout the region.
In the 1990s and early 2000s, international migration trends in CE countries remained highly diversifi ed, not only because each country had its peculiarities in terms of the scale and directions of mobility, but also because -and this is a point largely underestimated in existing research -the fi rst post-communist years were rich in different types of mobility, from cross-border commuting and petty trade to seasonal and short-term moves, to long-term migration for settlement purposes (Iglicka/Sword 1999;Sert 2018).The latter included not only the typical labour migration but also the migration of ethnic minorities, who in most cases were not allowed to enter the countries of their ancestors throughout the (entire or late) communist period: Germans in Hungary, Poland and Romania; Hungarians in Romania; Turks in Bulgaria (Gheorghiu 2005;Guentcheva et al. 2003;Korfalı/Acar 2018;Sandu et al. 2004).In turn, after the three Baltic states became independent, a signifi cant part of the Russian minority left those countries.In general, however, in all CE countries outfl ows have exceeded infl ows and net migration has remained negative, a situation that has only been exacerbated by the eastward enlargements of the European Union in 2004, 2007 and 2013 and the subsequent opening of the labour markets in the western EU member states.Interestingly, the post-enlargement outfl ow affected only a few CE countries, notably Lithuania, Poland and Romania, and took the form of a large outfl ow that was concentrated in the fi rst few years after the EU enlargement (Fic et al. 2016;Kahanec et al. 2016;Van Mol/de Valk 2016).Recent years have seen a further turnaround in international migration trends in the region, with emigration becoming less frequent, immigration becoming an integral part of some EU-11 labour markets, and an unprecedented increase in infl ows from non-European countries.Some EU-11 countries have started to record positive net migration, while the largest country in the region -Poland -has for several years been the EU's top issuer of fi rst residence permits to third-country nationals (Górny/ Kaczmarczyk 2019;Kaczmarczyk et al. 2020;OECD 2021).
This paper aims to focus on the determinants of emigration from CE countries from 1989 to recent years and to interpret the results of existing studies in the light of theoretical approaches to international mobility. 1We focus on migration for employment purposes, which dominates other motives of international mobility.Reference is made to several theories that are useful in explaining the determinants of outfl ow from the EU-11, as we believe that no single approach adequately explains the determinants of emigration from post-communist transition countries (and here we are in full agreement with Massey et al. 1999).However, we suggest considering the aspirations/capabilities approach as an overarching conceptual framework to interpret migration-related post-2004 realities (Czaika/Vothknecht 2014;de Haas 2021).In what follows, we present the main stipulations of migration theories and we incorporate them into the aspirations/capabilities approach.In the subsequent step, we theoretically investigate these stipulations and propose research questions that are adequate to the emigration from this region of Europe that took place during the post-communist transition, concerning in particular 1) the main motives for emigration and 2) the factors that have determined the scale, direction and temporal pattern (permanent versus temporary) of emigration.We address these questions by presenting and discussing existing empirical studies that have been carried out for CE countries since the 1990s.We aim to contribute to the literature by providing a broad interpretation of post-2004 mobility and pointing to commonly overlooked explanatory factors.To this end, the proposed conceptual framework (aspirations/ capabilities approach) is particularly useful as it highlights the role of mobilityrelated aspirations and the aspiration gap.
We begin by outlining the conceptual background to the analysis that will follow, in particular linking the aspirations/capabilities approach to the available theoretical frameworks.The next two sections examine capabilities and aspirations as explanatory factors.The fi nal section presents the conclusions to be drawn from the study.

Conceptual background
The determinants of migration have become the main topic of theoretical refl ection on the causes and mechanisms that intensify the spatial mobility of individuals, families and social groups (Zlotnik 1998).The variety of forms of migration, from involuntary fl ows as a result of political events or natural disasters to migrations that gradually intensify in the context of demographic growth or economic development, to short-distance cross-border mobility, implies a rich spectrum of determinants operating at different social levels.Moreover, migratory fl ows are the result of different motives and different external factors; there is no such thing as migration resulting solely from, for example, differences in wage levels between the country of origin and the country of destination.Similarly, there is no such thing as migration resulting solely from family commitments, undertaken in isolation from economic calculations or formal constraints associated with crossing borders and legalising residence.At the same time, however, migration theories are based on the conviction that it is possible to fi nd regularities in this diversity, for example when there are more migrations from less developed countries or regions to more developed countries or regions than in the opposite direction; when people who change their place of residence appear to have different demographic, social or cultural characteristics from the non-mobile population; or when certain migratory fl ows are characterised by different degrees of permanence, the importance of links with either the country of origin or with the diaspora (starting with the pioneering work of Ernst G. Ravenstein (1885), which paved the way for many others who have tried to interpret migration in terms of "laws").These regularities provide the basis for the development of theoretical concepts which, through synthesis and their own conceptual "grids", explain the causal factors of migration and make it possible to predict future mobility.Following numerous authors (Castles/Miller 1993;de Haas et al. 2019;Hammar et al. 1997;Massey 1990;Massey et al. 1993Massey et al. , 1999)), we postulate that migration should be seen as a complex socio-economic phenomenon that takes place in specifi c historical circumstances, and that holistic conceptual approaches should be used to explain territorial mobility.To this end, we refer to the aspirations/capabilities approach proposed by Hein de Haas (2021Haas ( , 2010)), who draws attention to the fact that most migration phenomena result from the complex function of aspirations and capabilities.For the sake of argument, it should be noted that de Haas was inspired by Jorgen Carling (2002), who tried to explain the complexity of migration decisions in the case of Cape Verdeans and formulated the concept of "forced immobility".In de Haas's approach, aspirations are a set of individual goals and feasible plans that have their structural sources (e.g. the socialisation process, education, cultural infl uences) but can also result from access to information.Aspirations are not limited to economic aspects and can change over time.Opportunities, in turn, can be defi ned through the prism of the freedom to pursue personal goals and make choices, which in practice translates into access to specifi c resources, such as fi nancial or social capital, that allow for the realisation of this agency.Mathias Czaika and Marc Vothknecht (2014) go one step further, conceptualising migration as a function of an individual's capability for migration defi ned as a combination of two capacities: the capacity to aspire and the capacity to realise.In this approach, people are not merely income maximizers, but consider a variety of factors in their mobility decisions, including the aspirational gap, which is the difference between a person's current level of wealth/lifestyle/occupational position and the level they aspire to.Importantly, Czaika and Vothknecht (2014) have argued that aspirational capacity, which includes not only the ability to strive but also the knowledge of how to achieve a particular goal, can change over time and be inherited, but can also be a product of one's social environment (e.g.college, family home).
The aspiration gap that drives migration can arise both from changes in the sphere of possibilities to realise certain plans (e.g. less restrictive migration policies, increase in individual competencies) and from growing aspirations and expectations; for example, due to the acquisition of a higher level of education.Thanks to such a two-sided perspective, the aspirations/capabilities approach provides a suffi ciently comprehensive framework to encompass and organise the full range of migration determinants that have previously been studied separately in different theories of international mobility.For example, large disparities in income and living standards between the countries of Central Europe and destination countries have contributed to the emergence of labour migration potential; that is, groups of people aspiring to improve their well-being through labour migration (Fig. 1).This motive for emigration is usually addressed by neoclassical economic theory.However, the emigration potential would never have been transformed into a massive permanent outfl ow if the CE countries had not become members of the European Union (EU), which increased the migratory capabilities of the citizens of CE countries.This factor, explained by world systems theory, had an important impact on the pattern of emigration after the 2004, 2007 and 2013 EU enlargements: its scale, geographical directions and temporal character.Obviously, enlargements have helped to reduce the costs of migration and, as such, have also been of interest to proponents of the neoclassical approach.Indeed, each factor contributing to migration, such as the end of mobility restrictions in the enlarged EU, can be interpreted by more than one theoretical approach.In Figure 1, we propose a catalogue of different determinants and factors that have shaped emigration from CE countries since 1989, as well as the migration theories that have been previously used to investigate each of them (Brettel/Hollifi eld 2022; de Haas 2021; Hammar et al. 1997;Massey et al. 1993;Zlotnik 1998).In the following, we present each of the listed theories together with empirical studies that have investigated these determinants of emigration from CE countries.
By encompassing different motives for emigration and the structural factors that enable these motives to be realised, the theoretical approach proposed by de Haas Fig. 1 corresponds to our above-stated intention to study migration in a complex and holistic way.By no means do we claim that the aspirations/capabilities approach has better explanatory properties than any other migration theory.However, there is one important implication of the aspirations/capabilities approach that needs to be emphasised: Aspirations provide a hitherto missing link between external (structural) conditions, their perception, and individual causation, the latter constituting a real possibility of fulfi lment of migration aspirations.

Income and living-standard disparities -the neoclassical economic theory
As early as the late eighteenth century, Adam Smith (1776Smith ( /2000) ) noted that income differentials provided an obvious basis for "arbitrage" in the form of labour mobility, while also writing about the unique resistance of man to movement.More than 150 years later, John Hicks (1932) wrote that the main impetus for migration is a disparity in economic benefi ts, especially in wage levels.These statements sum up the neoclassical economic approach, which sees human simply as a mechanism for rebalancing labour markets.In fact, in the theory of foreign trade, the Heckscher-Ohlin-Samuelson model concerns two countries with different endowments of labour and capital (Samuelson 1953).The removal of trade barriers leads to economic specialisation and the exchange of goods between the countries, and the country that is better endowed with labour (relative to capital) will be characterised by low wage levels and will specialise in the production of labour-intensive goods, which will be exchanged for capital-intensive goods through foreign trade.In the long run, wage differentials between countries may even out through trade exchange alone.
Taking into account labour migration completes the picture: workers will move to countries offering higher wage levels, while capital will fl ow to places offering higher interest rates.Thus, if the assumptions of the model are accepted, it turns out that three equivalent processes equalise wage inequalities at the global level: foreign trade, labour mobility and capital fl ows (direct investment).Consequently, trade liberalisation can impact migration and, in particular, should signifi cantly reduce it.
The relationships described by this general model become much more complex if we take into account aspects such as limited factor mobility, transport costs, differences in market structures, or technological progress.Developers of contemporary macroeconomic models successfully incorporate these elements into their methodological approaches (van den Berg/Bodvarsson 2009), but their conclusions are in some cases opposite to those described above, for example when trade liberalisation generates additional migration fl ows.The neoclassical macroeconomic models accurately indicate the general direction of factor fl ows and successfully explain the scale of potential migration, but they fail to explain real migration processes for several reasons.First, neoclassical models assume that people are extremely mobile, which means that mobility becomes the easiest way to redress inequalities on a global scale.Second, the models ignore the costs and barriers to migration or limit the analysis to fi nancial costs.And third, the models overlook the fact that migration makes no sense without a demand for workers and ignore the importance of the demand for foreign labour.This approach instrumentalises mobile workers and focuses on the concept of market equilibrium, the absence of which is supposed to be the main causal factor of mobility.
Based on the implications of neoclassical economic theory, the obvious research questions about emigration from Central Europe are as follows: Q1: Were the income differentials (an aspirational factor) the main driver of emigration from the CE countries?

Q2: To what extent did emigration from the CE countries constitute an arbitrage between labour markets, leading to a levelling of the income and living-conditions disparities?
Q3: Was the increase in capital fl ows between the "old" and "new" EU member states causally related, positively or negatively, to the increase in labour migration from the latter to the former?
Many empirical studies have addressed Q1, highlighting the role of income differentials in the intensifi cation of labour migration from CE countries to (main) destinations, in particular the western part of the European continent (Kahanec et al. 2009).At the micro-social level, the economic gains were even higher for migrants undertaking temporary mobility under bilateral agreements between the countries of origin and Western countries (Mansoor/Quillin 2006).In this case, migrants' earnings (and remittances) were spent in the country of origin and migrants benefi ted from the price-level differences.However, there are at least two phenomena which show that the relationship between income levels and labour mobility, as interpreted at the macroeconomic level, is not so straightforward.First, the scale of emigration continued to vary considerably during the period under review, with the Czech Republic, Hungary and Slovenia recording relatively low numbers of emigrants compared to other CE countries, despite similar wage levels and living standards in this region (Górny/Kaczmarczyk 2019).Labour market imbalances related to the post-communist restructuring process in the sending countries, in particular occupational mismatches and unemployment levels, were found to be crucial in stimulating the emigration potential (Kaczmarczyk/Okólski 2008; Kureková 2011; Tyrowicz/van der Velde 2018).Second, similar differences in emigration levels were found in certain regions which are not necessarily the least developed, but which have intense traditions of labour mobility.This was particularly true for border regions with ethnic minorities, such as the western part of Romania (Anghel 2016;Fox 2003) or Poland (Łukaniszyn-Domaszewska/Jończy 2021).Last but not least, how risky it is to apply the basic neoclassical approach (and the wage-differential argument) to complex migration processes, particularly in the CE context, is illustrated by the seminal work of Layard et al. (1992).
Beyond the short-term effect, CE emigration has not contributed to a reduction in income levels in the destination countries and therefore has not constituted an arbitrage between the EU labour markets (Q2) (Bruecker et al. 2009;Lemos/Portes 2008).Some effects have been observed for selected groups of the population: an apparent increase in income for the migrants themselves and their relatives receiving remittances (Mansoor/Quillin 2006), and a decrease in wages for medium-skilled workers, who are most exposed to increased competition in the labour markets of destination countries (Blanchfl ower/Shadforth 2009; Bruecker et al. 2009).In the sending countries, the increase in incomes has been attributed to economic restructuring, rather than emigration (Rutkowski 2007) and was relatively limited (Holland et al. 2011).Most studies show that convergence is relatively limited and should be attributed to factors other than cross-border mobility, and mainly to investment and structural change (Alcidi 2019;Galgoczi 2019).Indeed, economic growth in the CE countries remained the main factor behind the increase in income levels, as it was also the main driver of capital fl ows between the "old" and "new" EU member states (Q3).The increase in foreign direct investment (FDI) was observed long before the EU enlargements, already in the early 1990s, thanks to "the general perspective of improved circumstances" of investment (Breitenfellner et al. 2008: 114).Due to the Structural and Cohesion Funds of the EU, the FDI even increased after the EU's eastward enlargements (Breuss et al. 2010), and remains unrelated to the labour migration from the "new" EU member states, at least in the short term.

Improvement in education levels -the neoclassical economic theory
At the micro level, neoclassical economic theory assumes that individuals make rational decisions about the allocation of scarce resources, to maximise their utility.Individuals are expected to adopt a rational strategy of investing in their own human capital (education, training, etc.).Extending the time spent learning and studying, and acquiring rare skills and competencies in demand on the labour market, serves to increase future earnings and the rate of return to education.In this interpretation, increasing individual human capital is an investment that primarily incurs costs in the present and yields benefi ts in the future.Similarly, labour migration can be interpreted as a way of allocating human resources to increase their productivity.In the approach of human capital investment, based on the pioneering work of Larry Sjaastad (1962), each individual compares the benefi ts and costs (including non-fi nancial costs such as psychological stress) of working in the country of origin and abroad, considers the probabilities of certain events such as fi nding a job, being unemployed, being employed on a regular or irregular basis, and, based on his or her preferences for current and future consumption, undertakes labour migration whenever it is more profi table than not moving.This model has been criticised for several reasons, the main one being the assumption that individuals act in a highly individualistic and autonomous manner and do not take into account the preferences of their family members.Moreover, in the basic version of this approach (and this is commonly quoted), migration does not involve any costs or risks associated with fi nding a job, and individuals have access to perfect information about job opportunities abroad.
Nevertheless, in neoclassical economics, individuals are expected to compare their country of origin with other countries (international labour markets) and decide to migrate if the return on investment in human capital is higher abroad.The higher cost of education is therefore likely to lead to a search for the highest possible returns to education, which are mainly offered by highly developed countries.In the de Haas-inspired framework, this translates into higher aspirations for individuals who have invested in their skills and competencies.The institutional transition to a capitalist economy has contributed to the emergence of new professions and jobs in the private sector that offer high returns to university degrees, foreign language skills and other competencies useful for working in an international environment.This, in turn, has led to a real revolution in terms of human capital formation: in the CE countries as a whole, according to the European Labour Force Survey, the share of university graduates in the total population rose from 15 percent in 2000 to 36 percent in 2020.But as the number of highly educated people has increased in a relatively short period, especially among younger cohorts, so has the number of people potentially interested in labour migration.This leads us to the following research questions: Q4: Were highly educated people more likely to emigrate than those with lower levels of education?Q5: Was the rate of return to education higher abroad than in the country of origin?
In the case of most CE countries, there was a clear pattern of positive selection of persons with tertiary education among emigrants, especially since the accession of these countries to the EU (Drinkwater et al. 2009;Eade et al. 2006;Grabowska-Lusińska/Okólski 2009;Hazans/Philips 2010;Jakubiak 2015;Kaczmarczyk/Tyrowicz 2015;Kahanec et al. 2009).In other words, even when controlling for the age structure of emigrants and non-migrants, the highly skilled were overrepresented in the group of emigrants compared to the total sending population.Since the early post-accession years, this issue has been at the centre of academic debate and there is general agreement that the potential negative impact of such a brain drain depends on the structural circumstances of a particular sending country and the composition of the emigrants.Theoretically, the emigration of highly educated or skilled individuals has a negative impact on labour supply and hence on the economic performance of a sending country.However, several studies have interpreted post-accession migration in the CE region as a result of the general labour market situation and an oversupply of educated professionals whose skills could not be matched with job offers in the sending countries (Kahanec et al. 2009;White et al. 2018), while other research pointed to a very limited scale of (theoretically possible) brain gain (Bruecker et al. 2009;Kaczmarczyk/Tyrowicz 2015;Kaczmarczyk et al. 2020).
Regarding the last point, a large number of studies showed that the rate of return to education abroad was not necessarily higher than in the countries of origin, mostly due to large mismatches between the skills of migrants and the occupational position they secured in the main destination countries (the phenomenon of brain waste).These mismatches were only partly due to the (poor) quality of education in countries of origin or the lack of transferability of skills between countries.Rather, the available empirical studies point to the role of the structure of demand for foreign labour, which is heavily concentrated at the low-skilled end of the occupational hierarchy (Bruecker et al. 2009).The so-called brain waste phenomenon has been confi rmed for the main destination countries, including the UK, Germany, Italy and Norway (Bruecker et al. 2009;Galgoczi et al. 2009).Importantly, postaccession migrants from CE countries seem to fare considerably worse in terms of skill-occupation matching than migrants from pre-enlargement cohorts and do not improve their labour market position despite gradually better language skills (Kahanec/Zimmermann 2016; Voitchovsky 2014).Moreover, these effects were the strongest for the best educated migrants (Kaczmarczyk/Tyrowicz 2015).Thus, some authors (Galgoczi et al. 2009) have argued that post-accession CE migrants are mainly attracted by absolute wage differentials and behave as typical wage earners who remain focused on accumulating savings or remittances.

Market failures and risk diversifi cation -the New Economics of Labour Migration
The so-called New Economics of Labour Migration (NELM), put forward by Oded Stark and David Bloom (1985), challenged the individual nature of migration decisions.Although based on microeconomics, the approach successfully deepened the understanding of contemporary migration by incorporating elements of social psychology, sociology and household theory.The authors argued that labour, as a factor of production, is different from other factors of production and therefore requires its own methods of analysis.Furthermore, they argued that migration need not be permanent; on the contrary, temporary and circular movements are very common in the contemporary world; and fi nally, that people operate within social reference systems that should be considered not only as a context for migration decisions but also as a valuable aspect of interpretation.
The authors of NELM focused on labour migration from less developed countries that are characterised by market failures; for example, in the absence of an insurance system against natural disasters or a welfare system that provides fair unemployment benefi ts, people depend solely on themselves and their family members.These market imperfections accumulate various types of uncertainty about production and consumption, which in turn leads people to diversify risk, primarily in terms of sources of income (Stark/Levhari 1982).The novelty of this approach lies in the consideration that individuals do not act in isolation, but take important social and economic decisions within larger groups, usually families or households.In less developed countries, labour remains the most valuable resource owned by households, and the allocation of labour can be an effective diversifi cation measure to reduce the risk associated with daily income.Individuals, guided by intuition, experience, or resources accumulated in social capital networks, make decisions that do not maximise their utility (as neoclassical economists would suggest), but that secure the interests of the whole household and minimise the level of risk associated with income.A direct expression of this type of motivation is the remittances sent by migrants to their families, which in many cases not only maintain the consumption level of the part of the household remaining in the country of origin but can also be a source of investment (Amuedo-Dorantes/ Pozo 2011Pozo , 2006;;Osili 2007).Based on these considerations, the following research question can be formulated: To what extent has international labour mobility from CE been a strategy for households to diversify risks related to income sources?
At the country level, low social spending and inadequate welfare-state institutions in the countries of origin have been identifi ed as factors increasing the propensity to emigrate (Kureková 2013).However, at the micro-social level, households' risk diversifi cation strategies have not been the subject of any studies.Nevertheless, the characteristics of some migratory fl ows, especially temporary ones, clearly point to labour market failures in the countries of origin.A study on the seasonal mobility of Polish citizens to German agriculture, a fl ow that at its peak included 300,000 people per year, examined the economic strategies of temporary migrants who, in times of severe restructuring, labour market reforms, and high unemployment, combined the opportunities of temporary work abroad with life in the country of origin (Fihel/ Grabowska-Lusińska 2014).The study showed that this type of migration usually involved one adult member of a household in Poland and was repeated, once a year, for a relatively short period.Such a strategy was used not only by unemployed or economically inactive household members, such as housewives raising small children, but also by fully employed persons in occupations such as farming, driving, and teaching, who migrated during annual holidays or unpaid leave.Earnings from temporary employment in Germany were not the main source of household income, but they secured the well-being of whole families.Most of the available studies on CE show the complexity of the links between remittances and the development of the country of origin; on the one hand, there is evidence of a positive albeit small impact in terms of reducing inequalities, on the other hand, the structural and longterm impact is much more doubtful, mostly due to structural limitations at the point of origin (Kaczmarczyk et al. 2020;Vadean et al. 2019; Vullnetari/King 2011).

Increasing socioeconomic disparities -the New Economics of Labour Migration
One of the key concepts of the NELM is relative deprivation, which occurs when the utility of a given person depends not only on one's income but also on the income of people belonging to a so-called reference group.The reference group includes relevant people with whom an individual makes comparisons, such as relatives, neighbours, and local community members.Increasing one's income, for example through labour mobility, is a consequence of long-term dissatisfaction with one's social position within the reference group and is aimed at changing this position (Stark 1984(Stark , 1991)).This approach is extremely useful for studying temporary labour mobility, which consists of engaging in economic activity abroad while maintaining a relatively stable reference group in the country of origin.Such migrants are willing to work below their skills abroad as long as their position in the home reference group, as defi ned by income and consumption, remains high.Moreover, temporary migration can be seen as a way of avoiding relative deprivation.While relative deprivation does not apply to all types of labour migration, this phenomenon explains mobility in countries where socioeconomic inequalities are increasing rapidly (Czaika/de Haas 2012).This was the case in the CE countries undergoing post-communist transition (Izyumov 2010) where seasonal labour mobility provided an opportunity to earn higher incomes than in the place of living in a relatively short period.A corresponding research question is therefore proposed:

Q7: Has the international labour mobility from CE countries been driven by increasing socio-economic disparities in the countries of origin?
The empirical evidence is mixed in this regard.On the one hand, a study carried out for most countries in the world shows a weak and ambiguous effect of relative deprivation in countries of origin on actual emigration (Czaika/de Haas 2012).In a study of post-communist countries, emigration tended to decline throughout the pre-accession period, despite growing economic disparities between regions (Fidrmuc 2004).However, this result may be explained by the temporary nature of the outfl ow from CE, which was suddenly triggered by institutional reforms, in particular the introduction of visa-free travel to European countries.On the other hand, increasing income disparities between regions in Poland have been found to increase the propensity to emigrate, even when controlling for the absolute level of per capita income (Stark et al. 2009).For several EU-11 countries, high levels of inequality were found to increase outfl ows to the "old" EU member states characterised by lower levels of inequality, although absolute levels of income were not controlled for (Plotnikova/Ulceluse 2022) and no single study allowed also for the proneness of the low socio-economic groups to undertake labour mobility.It is important to note that the case of CE and post-2004 migration illustrates the role of different levels of relative deprivation, as this phenomenon can relate not only to income as such, but also to social position, role in the occupational hierarchy, etc.Many of these have been described as important parts of the aspiration gap, particularly for young migrants (Grabowska 2019;Grabowska et al. 2017;Trevena 2013;White et al. 2018).

4
Capabilities as determinants of CE emigration

End of mobility restrictions and the EU enlargement -World systems theory
World systems theory emphasises the importance of the uneven development of economies and political systems, which perpetuates dependencies between countries and world regions.The author of the approach, Immanuel Wallerstein (2011), aimed to explore the origins and dynamics of capitalism as an economic system that has largely shaped our contemporary world.The concept of the world system refers to the international division of labour, which is relatively permanent, though of course not immutable, structure made up of core, semi-periphery and periphery.While the core plays a dominant economic and political role and seeks the subordination of other areas, the periphery supplies labour and raw materials and is also an important market for the core's production.The semi-periphery, on the other hand, comprises all the entities that simultaneously "exploit" the periphery and are "exploited" by the core.
The spread of capitalism from the core to other regions of the world has serious consequences for the functioning of pre-capitalist communities: it leads to the breakdown of traditional land tenure systems and consequently undermines previously developed ways of farming and securing livelihoods.The increased demand for labour in extractive industries, accompanied by the gradual displacement of people from the natural rural economy, leads to the formation and consolidation of labour markets.It also contributes to the disruption of traditional social and economic relations, uprooting and creating a potentially mobile population able, and sometimes forced, to seek livelihoods outside their place of origin.In a system of dependency, migration does not necessarily result from wage differentials, but rather from the dynamics of economic systems in which the core is characterised by a permanent surplus of labour demand and the periphery constitutes a natural reservoir of the workforce.For obvious reasons, differences in levels of development determine not only the direction and volume of fl ows but also their structure; that is, the skills and competencies of migrants: mobile inhabitants of the core are mainly specialists, whereas migrating inhabitants of the periphery are both unskilled workers who undertake employment that the citizens of the core are not interested in, and specialists who undertake jobs in accordance with their qualifi cations.
The eastward EU enlargements and the integration of the CE countries into the core of the world system (i.e. the EU) have had far-reaching consequences for the ability of EU-11 nationals to move abroad.We therefore propose the following research question:

Q8: To what extent have the transitory labour market restrictions introduced during the EU enlargements determined the geographical directions of emigration from the EU-11?
Several studies have shown that the timing of the transitional labour market restrictions, which was not identical in all "old" EU Member States, infl uenced the geographical direction of fl ows.In the two decades that have passed since the fi rst EU enlargement ( 2004), the number of CE citizens has risen across the EU, but this increase was neither uniform nor proportional to the size of the labour markets (Van Mol/de Valk 2016).In 2004, the massive outfl ow from the EU-11 was to Ireland and the UK, the two out of three countries that did not introduce any transitory restrictions (Fihel/Okólski 2009;Okólski/Salt 2014).By attracting large numbers of highly skilled migrants eager to join the core of the global economy, these countries "won" the human capital, while others, such as Germany, missed this opportunity (Elsner/Zimmermann 2016).The transitory restrictions disadvantaged the labour market performance of EU-11 migrants; in particular the employment rates and the extent to which employment matched the qualifi cations (García-Gómez et al. 2021).At the same time, these institutional barriers did not prevent the presence of EU-11 workers who provided their services as self-employed or delegated employees.The case of the (removal of) restrictions on the labour market and its role in the migration after 2004 has been the subject of an analysis presented by Paweł Kaczmarczyk and Marek Okólski (2008).Their argument is that structural mismatches in the pre-2004 period were responsible for a massive migration potential which, however, did not translate into real migration fl ows.It was only in the post-enlargement period that, for the fi rst time in Polish history, migration became an option for expelling labour surpluses (crowding-out migration) and, in turn, reallocating labour resources in the domestic labour market.From today's point of view, the take-up of this opportunity has been limited so far (Kaczmarczyk et al. 2020).

Labour demand in destination countries -Theory of the dual labour market
Most theories explaining the determinants of international migration focus either on macro-structural factors that redress economic or demographic imbalances or on the motives of migrants to satisfy their needs.The approach put forward in the 1970s by Peter Doeringer and Michael Piore points to structural features of labour markets that make pure market mechanisms, as described by neoclassical economics, applicable only to part of the labour force: unskilled workers or those whose recruitment, training and retention costs are relatively low.A few years later, Piore (1979) went one step further and postulated that the labour market should be viewed through a sectoral prism as a dual system consisting of two basic structural types: the primary and the secondary sector.The former is characterised by high wages, good working conditions, employment stability with ample opportunities for career advancement and, at the same time, clear rules of the game in the form of administrative procedures.The latter is characterised by low wages, unfavourable working conditions, low job stability with few opportunities for career advancement (e.g.dead-end jobs) and often no rules or procedures.Such a picture of the labour market is simplistic and does not refl ect its complexity, but we could easily identify sectors featuring these characteristics: the primary sector including high technologies and tech services, as well as public administration, or the secondary sector including the seasonal work sectors (agriculture, catering, hotels and restaurants), manual services (e.g.domestic services), and even the entire range of illegal activities.
According to Piore, employment in the primary sector should be the goal and ambition of most workers, except for those social groups for whom employment in the secondary sector is either a necessity (e.g. the least educated) or not part of a long-term career (young people, women, farmers).However, due to several factors (in particular the entry of women into the primary labour market, the trend towards longer education, and the gradual decline of farmers as a professional and social group), this situation began to change in the post-war period, resulting in a permanent shortage of labour supply in secondary sector jobs.Employers sought to solve the problem by raising wages or trying to replace labour with capital (automation), but these were only moderately effective or even counterproductive.Thus employers in most developed countries began to look for cheap labour in less developed countries (the periphery), which means, according to Piore, that most contemporary labour migration is determined by the labour demand inherent in highly developed countries.Moreover, much of it has its origins in institutionalised recruitment, in some cases initiated by government agencies.For migrants, the secondary sector is often the only employment option available, while the low wages and poor working conditions may still be far more favourable than in the (peripheral) countries of origin.In addition, as long as immigrants consider their stay in a highly developed country to be temporary, they are willing to accept jobs with low social prestige, at the bottom of the occupational hierarchy.This is because such workers may treat work purely instrumentally; not as a mechanism for accumulating status and social capital, but only as a means of generating income that can be transferred to the country of origin and used there to establish social relations.
Based on the dual labour market theory, we propose the following research question:

Q9: In what way did the labour demand in the main destinations determine the scale and geographical directions of emigration from the EU-11?
This research question has been investigated in many studies, especially concerning post-accession migration, which turned out to have more of a settlement character than previous fl ows.The existence of a secondary sector in the labour market of the United Kingdom proved crucial in attracting large numbers of EU-11 migrants who found employment mostly in low-skilled occupations and such sectors as production, manufacturing, retail trade and hospitality (Drinkwater et al. 2009;Holland et al. 2011;Kahanec et al. 2009).The relatively favourable education structure among the migrants led to the phenomenon of brain waste (Bruecker et al. 2009;Kaczmarczyk/Tyrowicz 2015).Similarly, the secondary sectors in Italy and Spain and the widespread tolerance of irregular employment in these countries signifi cantly attracted Romanian migrants long before Romania's accession to the EU (Del Boca /Venturini 2016;Peixoto et al. 2012).In the Italian labour market, the main employment sectors for CE migrants are construction, hospitality and household services (Del Boca /Venturini 2016;Holland et al. 2011).Sweden, which, like the UK, did not introduce transitory labour market restrictions for EU nationals at the time of the 2004 EU enlargement, has not experienced signifi cant immigration, due to the absence or small size of their secondary sector of the labour market (Fihel/Okólski 2009).In a number of countries, the structure of demand (as well as the modes of recruitment) has been mainly responsible for specifi c forms of migration, i.e. highly masculinised seasonal labour mobility (Engbersen et al. 2010;Friberg 2012;Friberg et al. 2014).

Migration networks -Social capital theory
The application of social capital theory to migration research points to the role of social ties in facilitating and sustaining international mobility.So-called migration networks encompass social connections of migrants or migrants-to-be with relatives, acquaintances, neighbours, community members etc. who provide important information and resources, both material and non-material, throughout the migration (Gurak/Cases 1992).These ties can take different forms, from informal ties with family members and friends who have already settled abroad or know other migrants, to formal institutions, such as associations based in the countries of origin and destination.By providing key information on migration and employment opportunities, social networks reduce the risks associated with moving to another country and help migrants integrate with the host society.Migration networks can also infl uence the decisions of those potentially interested in undertaking international mobility, thus consolidating fl ows between two or more countries and making migration a self-perpetuating process.In the age of modern transport and communication technologies, migration networks create so-called transnational social spaces (Faist 2000) that link immigrant communities with their countries of origin and other countries.In addition, migrants' networks can be seen and interpreted as a kind of social infrastructure that reduces the risks and limits the costs and thus makes mobility easier and more accessible (i.e. also less selective; Guilmoto/Sandron 2001).As migration networks are diffi cult to operationalise and quantify, most studies focus on the functional and contextual aspects of their functioning.Therefore we propose the following research question in the context of post-communist CE countries:

Q10: In what ways have migration networks been useful for emigrants from the EU-11 countries?
Much of the research on this topic has focused on post-accession emigration from the EU-11, apparently due to the massive scale of the outfl ow and the growing interest in the phenomenon.However, pioneering ethnosurvey studies carried out in Poland in the 1990s and the early 2000s should be cited fi rst.Before the eastward enlargements of the EU, Western European labour markets were restricted to CE citizens, but the visa-free travel agreement allowed them to stay in many European countries for up to three months at a time.As a result, thousands of Polish citizens seeking employment in Western Europe undertook temporary mobility and took up illegal employment, mainly in the construction or domestic service sectors.Studies conducted in both the sending and host communities have shown that during that period of political transformation, strong migration networks had already developed, linking the Polish periphery -small towns and villages in underdeveloped regions -with Western European metropolises: Berlin, Brussels, London or Vienna (Jaźwińska/Okólski 2001;Kaczmarczyk/Łukowski 2004).Communities of temporary economic migrants emerged who regularly shuttled every few weeks between their homes in Poland and work in Western European capitals.In this case, the usefulness of migration networks was to circumvent the institutional barriers that existed in the pre-accession period in Western European countries and to provide important information and resources related to employment in the informal economy.Similar results were found for the pre-accession migration of Romanian nationals going to Spain: arriving under the guise of tourism, Romanian citizens "overstayed" their tourist visas and took up irregular employment with the help of friends and relatives (Serban/Voicu 2010).Migration networks helped new arrivals to circumvent restrictive immigration policies in Spain (Elrick/Ciobanu 2009) and other southern European countries (Ciobanu 2015), which became destinations for CE migrants due to their relatively high tolerance of irregular employment and the intense demand in the secondary sector of their labour markets.
The functional aspect of migration networks has changed since the EU enlargements, as the citizens of eleven CE countries are no longer or only temporarily subject to labour market restrictions.Migrants' connections have become more diversifi ed and multifaceted, extending both within and across similar migrant groups (Ryan et al. 2008) and into transnational social spheres (Marcu 2022).They have been favouring economic and social integration of EU-11 emigrants into host societies (Garapich 2008).Networking activities have been benefi cial for different types of migrants, from highly mobile professionals (Marcu 2022) to low-skilled people with poor skills in the language of the host society (Pavličková/Dayanandan 2015), but their use is still very much dependent on levels of trust towards compatriots (Schwabe/Węziak-Białowolska 2022).Moreover, despite the absence of labour market restrictions and the development of communication technologies, migration networks in the enlarged EU have remained a major channel through which employers search for migrant workers, whether high-or low-skilled (McCollum et al. 2013;McCollum/Apsite-Berina 2015).A study of the outfl ow from Latvia shows that these informal links between migrants, potential migrants and employers in destination countries shape the nature of the fl ows, in particular their geographical scope and temporal pattern (McCollum et al. 2013).This issue has been explored in depth in the context of the migration industry, which we present next.

International recruitment -Migration industry approach
When recruiting foreign workers, employers are not necessarily motivated by the lower reservation wages accepted by immigrants, but rather by their fl exibility (Caviedes 2010).This crucial feature relates to the immediate availability of migrants and their adaptation to labour demand in higher-income countries.Labour demand changes with the seasons, economic cycles, and structural shifts, the latter creating pressing skill shortages.The study of Latvian migrants cited above (McCollum et al. 2013) focused on their quick responsiveness to the needs of employers in the EU, needs expressed through migration networks and other channels, in particular institutions of the so-called migration industry.
Migration industry is another approach, alongside the theory of the dual labour market, that focuses on the demand for foreign labour as a driver of mobility.This approach emphasises the relevance of all private actors and organisations that profi t from the intermediation between migrants and their employers, and between the sending community and the receiving society (Gammeltoft-Hansen/Sørensen 2013).This "infrastructure" includes, fi rst and foremost, recruitment agencies that match migrants or migrants-to-be with abundant employment opportunities abroad (Cohen 2022).International recruitment by external agencies can be legitimate and mutually benefi cial, but it can also be exploitative, especially when the responsibility of employers towards workers is blurred by profi t-driven intermediaries (Alberti et al. 2015).This relatively new area of research does not refer to any general social theory, but Sophie Cranston et al. (2018) distinguish three major strands of research, the fi rst of which, the structuralist approach, emphasises the interweaving of commercial intermediaries (brokers, recruiters, transporters, etc.) and informal actors (migration networks) -the former often emanating from the latter -in overcoming restrictive migration policies and border controls.Second, the labour market approach focuses on the role of labour market agents: temporary work agencies, contractors, and recruiters, in channelling information and resources to address structural labour shortages.Finally, the mobilities approach highlights migration as a process, a phenomenon of being "on the move", and investigates how different actors and networks facilitate these fl exible forms of mobility and various transnational practices.Paradoxically, despite the absence of labour market restrictions within the EU, the number of studies dedicated to the role of the migration industry in stimulating infl ows from the EU-11 countries has increased in recent years, but this may be due merely to the general growing interest in this area of research.Nevertheless, the theory-driven research question would be as follows:

Q11: How has the migration industry affected emigration from the EU-11 countries?
Several studies have addressed this question by pointing to specifi c sectors of the labour market, such as care and domestic services, which require the intermediation of recruiters and temporary work agencies, mainly due to the dispersion of employers.The role of the migration industry proved to be crucial in initiating and channelling the outfl ow from the EU-11 countries, addressing skill mismatches and contributing to a more effi cient allocation of labour in Germany (Leiber et al. 2019), the Netherlands (Da Roit/van Bochove 2017; Szytniewski/van der Haar 2022), Norway (Friberg et al. 2014;Žabko et al. 2018) and the UK (Sporton 2013), even in times of COVID-19-related mobility restrictions (Nowicka et al. 2021).Other studies pointed to the practices and rhetoric of labour market actors that infl uence migrants' identities (Findlay et al. 2013;Garapich 2008;Hopkins 2017; MacKenzie/ Forde 2009): in general, the EU-11 migrants were perceived by employers as "good workers" with a strong work ethic, neat appearance, and diligence in performing their duties, contributing to the persistence of labour migration from Central to Western European countries.

Conclusions
In this paper, we have attempted to pose eleven theory-driven questions about the main determinants of the mobility and emigration from the EU-11 taking place since 1989 and to fi nd answers to these questions in existing empirical studies.The large body of research dedicated to examining the outfl ows from the EU-11 shows that while wage differentials remain an important aspirational driver of intra-EU labour mobility (Q1), their impact is mediated by two other factors.First, the labour market imbalances in the sending countries, especially occupational mismatches and unemployment levels, have been leading migrants to adopt household-based strategies aimed at reducing risks related to income sources (Q6).In addition, the high level of unemployment in the period before 2004 suggests that instead of referring to "raw" wages, we should rather look at expected wages, which also take into account the probability of fi nding a job.At the same time, growing socio-economic disparities in the countries of origin have not necessarily affected the willingness to migrate (Q7).Second, the traditions of labour mobility have created lasting links between countries that have enhanced the capability to undertake migration: before the EU eastward enlargements they served to circumvent immigration restrictions; today they contribute in a variety of ways to the better economic and social integration of migrants (Q10).In the same way, the emerging migration industry has directed EU-11 nationals towards certain sectors of the labour market (Q11), as in some EU countries the secondary labour market sector has attracted EU-11 migrants (Q9) notwithstanding their relatively good competencies and skills (Q4).Thus the rate of return to education was not higher for all migrant groups in Western European countries than in the EU-11, as most of the available jobs taken by migrants were low-skilled and low-paid (Q5).Finally, the schedules of the EU enlargements and the transitory labour market restrictions infl uenced the geographical directions of mobility, with countries without immigration barriers "winning" highly skilled migrants (Q8).The levelling of the income and living conditions due to migration (Q2) was observed mostly for migrants and selected groups in the host societies, while in countries of origin, the increase in wages was not related to migration (or related to a limited extent only).So far, no correlation between labour and capital fl ows (Q3), which would indicate a substitution effect between the two, has been seen.
The selection of factors underlying the outward migration from the EU-11 presented in this paper is simplifi ed and limited to what we consider to be the most important, and could be a subject of discussion.One can think of other factors that have encouraged EU-11 nationals to migrate for work, in particular the decreasing transport costs and the improved access to labour market information following the EU enlargement (a capability), which is perfectly explained by human capital theory, or the formation of a middle class integrated into the core of the world system (an aspiration), which is the subject of world systems theory.Rather than enumerating all possible determinants of migration, however, we sought to test the explanatory power of the aspirations/capabilities approach, and we conclude that, while it serves well as a general framework, it also draws on other theories that focus on a single, selected factor underlying international mobility.Such a framework will remain useful in explaining other migration-related phenomena that are becoming increasingly important in the EU-11, in particular the recurrent mobility of highlyskilled workers, return migration, and the infl ow of third-country nationals.
Importantly, the analysis presented shows the full complexity of post-2004 CE migration.Despite the fact that most fl ows were for work reasons, they were driven by a variety of socio-economic factors.In view of this fact, as well as the great diversity of migration strategies/tactics, it should not be expected that this process could be easily explained and interpreted with reference to only one theoretical approach, and this applies in particular to the argument of wage differentials that is often invoked.The aspirations/capabilities approach emphasises the relevance of structural determinants of migration on the one hand, while on the other hand it believes that individual decisions are responsible for the selectivity of migration.The intermingling of these two perspectives overcomes, at least to some extent, the conceptual dichotomy between structures and agency as the main factor underlying international migration that is characteristic of other theories of migration (de Haas et al. 2019).Our study shows that the aspirations/capabilities framework has made it possible to illustrate the complexity of recent migration from CE and to highlight a large number of underlying factors.At the same time, the aspirations/capabilities framework is not ahistorical but highlights the role of specifi c circumstances, such as the post-communist transition to a capitalist economy and liberal political system, in increasing the aspiration gap and the propensity to migrate.As this is a review paper, there has been no attempt to do a dedicated empirical analysis of the issue.That would have revealed another limitation of the approach, in that many variables can be interpreted both as desired and as achieved.Migration from CE is a continuously evolving process and its structure and patterns in the early 2020s are signifi cantly different from those of 20 years ago, i.e. around the time of EU enlargement.Over the years, migration from the region has become more settlement-oriented and much more diverse in terms of its motives.Moreover -and this would be an additional research challenge -most CE countries have already become net immigration areas, which in our opinion calls for an in-depth analysis of the link between emigration and immigration (which was beyond the scope of this paper).