DIVERSIFICATION TRENDS IN MOLDOVAN INTERNATIONAL MIGRATION: EVIDENCE FROM CZECHIA AND ITALY

https://doi.org/10.14712/23361980.2017.19 Drbohlav, D. – Bailey, A. J. – Čermák, Z. – Čermáková, D. – Lozovanu, D. – Masná, E. – Pavelková, L. – Seidlová, M. – Stojanov, R. – Valenta, O. – Vietti, F. (2017): Diversification trends in Moldovan international migration: evidence from Czechia and Italy AUC Geographica, 52, No. 2, pp. 237–248 © 2017 The Authors. This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0). ABSTRACT


Introduction and main goals
Approximately 250 million international migrants -3.4% of the world's population -had been living outside their country of birth for more than 1 year in 2014 (World Bank 2016a). As most migrants move in the direction of economic opportunity the "migration-development nexus" has emerged as an influential explanation of, and policy heuristic for international migration (Faist 2004;King et al. 2013). Recent work contends that migration issues have a pendulum-like character, going back and forth from more pessimistic to more optimistic treatments as de Haas ( , 2012. For example, around the beginning of the new millennium, the overall assessment was often in positive tone, with financial remittances offering a winwin-win for origins, destinations, and migrants through processes of brain gain, positive role of diasporas, and circulation migration (GCIM 2005). However, securitization, global economic recession, persisting discrimination against migrants, the impact of circulation upon integration, and a wider breakdown of migrant social and family relations have suggested the pendulum may be swinging back ( de Haas 2007b;Triandafyllidou 2013). Set within our broad goal of advancing an understanding of contemporary international migration, the specific objective of this paper is to review evidence on how the concept of the migration-development nexus applies to the particular situation of Moldovan migration. The Moldovan migration system is a compelling one to study. Moldova is deeply affected by migration and has received financial remittances since independence (Ruggiero 2005;Pinger 2010). Financial remittances of Moldovan migrants represent a very important source of income (Piracha, Saraogi 2011;Siegel, Lücke 2013). As a result, Moldovans move globally, with new streams touching new origins and reception contexts, including southern Europe and Central and Eastern Europe (CEE). However, continuing research reveals complicated negative externalities. This includes the issue of children left behind (Salah 2008;Gassmann et al. 2013) and the transformation of social roles in a transnational context. This draws attention to the social dynamics within the Moldovan diaspora and to the interactions between immigrants and the host society (Babenco, Zago 2008). Besides the role of the Orthodox Church as one of the main actors in constituting the individual and collective identity of Moldovans abroad, researchers highlighted the crucial activity of cultural associations and the professional and non-professional folkloric groups (Piovesan 2012). Moldovan migration is also driven by changing international and national policy regulations, particularly since the global economic recession and rising security concerns.
To study how Moldovan migration is driven by these multilayered economic, political, and social processes we compare one national group (Moldovans) across two destinations rather than comparing different national groups in one destination (for example Marchetti and Venturini's analysis of Moldovans and Ukrainians in Italy, 2014). This helps explore how the different migrant practices, reception contexts, and policies of the migration-development nexus affect Moldovan migration.
To begin to fill this research gap we use a comparative design to focus on Moldovans in a long established destination where migration amnesties affect labour markets (Italy) and in a more recent destination (Czechia) with more reactionary migration regulations. Qualitative research on Moldovans in Italy explores the female care-chain that links Moldovan women to the Italian families where they work and at the same time to their own families waiting for them at home in Moldova (Mazzacurati 2005;Boccagni 2009). These studies analyse the cultural, social and economic behaviour of migrants in terms of transnational practices, discussing issues related to the different household strategies, the wellbeing of "left-behind" people (especially children and grandparents), the changes in the use of the money sent by migrants, the transformations in contemporary Moldovan society driven by emigrants' social remittances. Czechia is the most migratory attractive country (in absolute terms) of all the CEE countries and, as a member of the European Union (EU), hosts about half a million legally resident foreigners including Moldovans who began arriving after 1989 (ČSÚ 2015;Drbohlav et al. 2010).
The paper makes two broad contributions. First, we are the first to assemble and contribute new knowledge on Moldovan migration to Czechia and Italy. To date in Czechia, there is no study focusing on Moldovan migrants and their migratory and integration patterns. Thus, this contribution is, to some extent, filling a gap that exists in this area of research by using a unique database (e.g. on foreigners' employment in Czechia). Second, we raise a series of research questions that may extend the migration-development nexus and guide policy development.
The rest of the paper is structured in three parts. We introduce the main ways in which Moldovan international migration has diversified since independence in 1991. We then describe and compare the experiences of Moldovans in Italy and Czechia, with an emphasis on labour market, historical developments, and geographic diversities. The third section discusses how the comparison extends the concept of the migration-development nexus. The final section summarises our general contribution to understanding Moldovan migration and concludes with a call for more research on social remittances and for robust data.

Diversification of post 1991 international migration of Moldovans
In common with many contemporary territories, the seemingly neat cartographic boundaries of today's Moldova obscure complicated geopolitical and cultural traditions. That is, the Republic of Moldova includes the eastern part of historical Moldova, which was in the last two centuries successively administered by the Russian Empire , Romania (1918Romania ( -1940Romania ( /1941Romania ( -1944 and the USSR (1944USSR ( -1991. While Moldova has been an independent state since 1991, part of what appears under government jurisdiction is not under the control of the authorities, instead constituting a secessionist "Dniester Moldovan Republic" in the eastern part of the country. These recent and multi-layered transformations are accompanied by equally complex and historically variable migration systems. Here, we focus on the significant changes that are unfolding in the geographies of international migration since 1991. The economic collapse that followed the political one led to the loss of many jobs in industry and agriculture, wages that were insufficient to meet basic needs, and a range of social problems. While for the years 1989-1991 emigration balanced immigration, 1992 marked a turning point, and since then emigration has exceeded immigration. In 1994 the number of emigrants was already double that of immigrants (Mosneaga 1999: 70-71).
Since independence international migration has been overwhelmingly influenced by Moldova's socioeconomic problems. Most migrants are classified as labour migrants. Of this population, 56% are males and 44% females. Their pattern of destinations has undergone significant transformations over the past twenty five years ( de Zwager, Sintov 2014). Preceding independence, the regions of the former Soviet Union, especially those with oil related employment (including Khanty-Mansiysk and Yamalo-Nenets), were dominant destinations. Recently, emigration to Europe has grown fast, particularly to the Mediterranean region, where Italy and Portugal emerged as the most important destinations, followed by Spain, Greece and Cyprus, at least until early 2008. Ireland became attractive during the economic boom and attracted migrants in 2002-2009. France, Belgium and Germany have become more attractive in the last decade and the Scandinavian countries have also recorded a gradual growth of Moldovan migrants in recent years. Among post-communist countries, Czechia is the most attractive and has the largest concentration of Moldovans, followed by Poland and Slovenia. Seasonal migration, defined as migration for less than 9 months per year, involves an estimated 109,000 Moldovans or represents 3.3% of the total population. 81% of seasonal workers migrate to Russia, with Italy a distant second with 7%. Most recently, there is an increase in the secondary migration of Moldovans to the United Kingdom (UK) from countries of the EU where they were previously employed, such as Romania, Portugal, Spain, Greece or Italy. Indeed, migration geographies now extend beyond Europe. New destinations include the United Arab Emirates (UAE), Qatar and Kuwait, South Korea and Australia. In summary, recent data suggests that 56% (or 206,000) of Moldovan labour migrants are in Russia, 22% (or 81,000) in Italy, and between 2% and 3% (or 8,000 to 10,000) in each of France, Turkey and Portugal (de Zwager, Sintov 2014).
The migration system is also diversifying to include a range of forms of mobility. For example, 18,400 Moldovans are studying abroad, and 18,700 people have migrated for the purpose of family reunification. Return migration is also increasing. Between 1992 and 2013, 44,000 international migrants, or 1.3% of the total Moldovan population, returned to Moldova, affecting 3.2% of Moldovan households. Of this number, 54.5% returned in 2010-2013; 49% returned from Russia, 15% from Italy and 12% from Israel ( de Zwager, Sintov 2014).
Taken together, these changes mean that, in Moldovan society, up to one in three persons are directly impacted by some form of mobility: 38.6% of Moldovan households had one or more persons involved in international migration and/or geographical mobility (internal mobility), 29.4% of households had one, more or all members involved in international migration, either long-term or seasonal. 12.4% of the total population is in long-term international migration, including 11% classified as labour migrants.
However, while migration touches many Moldovans the precise patterns of emigration vary across the different districts and localities in Moldova. Thus, most people who have emigrated to Italy, Portugal, Spain, France, Britain and other EU countries come from central districts of Moldova, and are mostly ethnic Moldovans/Romanians. Moldovans from northern, eastern and southern Moldova go mostly to Russia, whereas the majority of Moldovan migrants from southern Moldova, particularly from Gagauzia, go to Turkey. Territorial correlations can be found also at the level of settlements. For example, many migrants in Ireland are from the town of Durleşti (Chisinau), in the UK from the villages of Costesti (Ialoveni) and Colibaşi (Cahul), in France from Corjeuti and Caracuşenii Vechi (Briceni). In Italy and Portugal they are from several localities from Hincesti, Straseni, Nisporeni, Anenii Noi districts, etc.
These trends in the diversification of Moldovan migration point to a combination of familiar and unique processes. A large part of migration is economically motivated. Moldova is one of the poorest countries in Europe. GDP per capita (purchasing power parity in USD) was 4,893 -128th out of 183 assessed countries of the world (World Bank 2016b). Similarly, Moldova has a poor position in regard to the more complex human development index (HDI) -it stood at 0.693 in 2014, thus ranking Moldova in 107th position out of 185 countries of the world (UNDP 2015). After 1990, the Moldovan economy went through major structural transformation. The share of agricultural production on GDP dropped from 42% to 10% between 1989 and 2012, while high employment in the sector was not greatly changed -standing at 25% of the economically active population in 2012 (Ghedrovici 2014).
This punishing socioeconomic context means that many Moldovans turned to migration so they could send financial remittances to provide sufficient earnings to support family. Indeed, these financial remittances 1 are important in the aggregate to the Moldovan economy. From the economic perspective, for many European and Central Asian (ECA) countries, remittances are the second most important source of external finance after foreign direct investment (FDI). In Moldova, remittances bring in foreign exchange equivalent to almost half of export earnings. However, official remittances figures tend to undercount the actual flows by the amounts sent through informal networks in most instances (Mansoor, Quillin 2007). Moldova has the highest ratio of remittances to GDP income of all European countries (26% -World Bank 2016c) and in fact is among the economies with the highest ratio of this indicator in the world.
The significance of remitting helps explain why some migrants are drawn to prevailing high-wage regions of Russia as well as the EU countries, the USA, Canada, Australia, South Korea, UAE, Qatar, and Turkey. The institutional importance of remitting to the economy also helps explain why Moldovan migration has been increasingly subject to policies enacted in the name of the migration-development nexus. This includes a growing range of visa, entrance, and citizenship policies that, generally, affect labour market accessibility and opportunities to remit (GCIM 2005). Thus, visa-free access to Russia, Ukraine, Belarus and Kazakhstan has accelerated migration to these countries. While Moldova is not a member of the EU it has been increasingly affected by European policies that have sought to manage the migration-development nexus as part of a broader geopolitical, socio-economic, and ideological project (Carrera 2007). In 2003, Moldova became part of the European Neighbourhood Policy (ENP) aimed to bring democracy, the rule of law, a respect for human rights, and social cohesion to Europe's eastern and southern neighbours (European Commission 2015). As an ex-Soviet republic Moldova is also a member, since 2009, of Europe's Eastern Partnership (EaP), a forum for discussion of mutual interests in trade and political development. These general high level processes and dialogues impact Moldovan migration because they tend to carry Europe's increasingly connected foreign policy agendas and border management strategies. For example, Europe's Global Approach to Migration (GAM) sought to manage migration in ways that could be of use to Member States and, through remittances, for origin countries (GCIM 2005). In Moldova, this included the signing of a Mobility Partnership which aimed to provide economic remittances, benefit the economy through the transfer back of skills, and secure foreign investments. Under GAM, Moldovan border control and migration management were aligned with European protocols, with FRONTEX assuming a greater role in border control and readmission agreements proposed to increase the volume of short-term and temporary migration between Moldova and European Member States. A series of geopolitical and economic events -including instability in the Middle East and North Africa, and global recession -transformed European policy (Van Houtum 2010). For Moldova this ended the ENP Action Plan in June 2014 and replaced it with an Association Agreement between the EU and Moldova. This includes the Deep and Comprehensive Free Trade Agreement (AA/DCFTA) and affords Moldovans with biometric passports the opportunity to travel, visafree, to Schengen countries. In 2014, 340,000 Moldovans had used such facilities (European Commission 2015).
These changing policies have greatly increased the range of strategies that are available to, and are being used by Moldovans seeking mobility. Here we briefly introduce seven modes which variously extend the range and flexibility of opportunity for Moldovan migrants wishing to work to remit. First, emigration can be enabled by obtaining a pre-departure employment contract. This was originally important in states such as Israel, Poland, and the UAE, but now extends to work in the UK, Ireland, Germany, and Canada, although primarily on the basis of EU citizenship. Second, Moldovans leave with short-term visas enabling entry to the EU, the USA or Israel. After the expiry of their visa, some seek to remain as undocumented migrants in the country where they work. This strategy appears more important for the period until the abolition of the Schengen visa regime for Moldovan citizens in 2014. Third, Moldovans migrate based on specific education, working holiday, and other travel programs. This includes student visas, and may carry the possibility of post-education employment in the destination. Fourth, departures have been organized outside the above protocols ("illegally") by smugglers and traffickers, especially in the years 1994-2004. Fifth, Moldovans leave with false identity documents, particularly those issued from Lithuania, Latvia, Romania, and Bulgaria. The destination was often the UK, Ireland, Italy, France or Spain. Sixth, Moldovans depart having first legally obtained citizenship of Romania, Bulgaria, Portugal or, increasingly, other European countries. This strategy started in 1996 and has accelerated since 2010. The advantages include being able to work freely in EU countries and being able to obtain rights in other states as well. Seventh, Moldovans obtain legal status while overseas, and may return and subsequently circulate. This includes those who lived in Canada (province of Quebec), Israel (Jewish ethnicity), Germany (German by blood line), Russia (via a program for returning compatriots), and the USA (via a green card). Other Moldovans have obtained legal status overseas through family reunification.
We also argue that while socio-economic conditions and changing policies underpin diverse mobility strategies, social and cultural factors enable and constrain the exact patterns of migration. First, as is evident in the choice of priority regions for those leaving Moldova, language heritage matters. Moldovans are generally bilingual, speaking a Romance language (Romanian) and a Slavic one (Russian). Migration to Russia and other countries with Slavic languages (such as Czechia) is strongly influenced by this, while the Romanian language opens easier access to Italy, Spain, Portugal and France.
Second, while many Moldovans do migrate to improve themselves, widespread poverty in Moldova means that family needs are as important and sometimes more important than strictly individual aspirations. This means that families and kin networks play a strong role in influencing the direction, timing, and consequences of migration strategies, with ongoing implications for the migration system. For example, because many migrants (men and women) often have children or parents who remain in Moldova, their absence has a major influence on their relatives, communities, and social relations. This includes the issue of "abandoned children" in Moldova: "The number of children left behind is high: more than 100,000 children, according to a 2012 UNICEF report. In 2011, one out of every five children in Moldova had a parent living abroad, while 10 percent had both parents abroad" (Yanovich 2015). With rapidly changing social relations, including the feminization of migration from Moldova (Vanore, Siegel 2015), institutions including the Orthodox Church more readily speak out and intervene in migration issues, further complicating a diversifying migration system.

Moldovan migrants in Italy and Czechia
This section further investigates the diversification of Moldovan international migration by focusing on the experiences of migrants in Italy and Czechia, with an emphasis on historical development, geographic organization, and labour market structure. We start by presenting the findings for the longer established and larger community in Italy and then compare this with Czechia. In both cases, we use a combination of established "official" sources and secondary sources, including our own experience of working in these locations. Triangulating across different sources has two advantages. First, statistical data on international migration are subject to systematic and gradual development, and are often far from being comprehensive. In the case of Czechia there is a paucity of relevant data on numbers, flows, types, motives, or preferences of international migrants 2 . For example, the data on employment of Moldovans (and all other foreign citizens) in Czechia is not available for 2012 and 2013 due to a collapse of the system of data collection and processing at the Ministry of Labour and Social Affairs in early 2012. Second, many Moldovans migrate as Romanians due to historic and linguistic connections and cultural and ethnic affinities, and the reality that Moldovans who have a Romanian passport and are registered as Romanian citizens in the EU have unrestricted access to the EU labour market.

Italy
Italy transitioned from a country of emigration to one of immigration after the 1990s as a significant number of immigrants, especially from the Balkans and the Mediterranean area arrived. Official statistics suggest that Moldovans were part of this shift, at first comprising a small and indeed invisible group of 7,000 people i.e., about 0.5% of the total of all foreign residents (ISTAT 2002). While quantitative and census data initially undercounted the community, ethnographic and qualitative research began to describe, particularly in some abandoned areas of the suburbs of the large cities, the appearance of buses going back and forth between Italy and the countries of Eastern Europe. Here, every week, Moldovan migrants who lived in Turin, Padua or Bologna came to send packages, gifts, money, letters and photographs to friends and relatives who remained in their country of origin. Thus, in the midst of washing machines, microwave ovens, packs of pasta and bottles of perfume, these parking spots become the starting point of a transnational story that continues to transform both Moldova and Italy (Vietti 2012).
Crucially, the 2002 sanatoria (regularization of undocumented immigrants) itself linked to the so-called "Bossi-Fini Act" led to a swift increase in the visibility of CEE communities. Women migrants were a key group to be recognized, of whom many had been employed as domestic workers or care assistants to elderly Italians. Italian society thus discovered the binomial badanti moldave as "Moldovan care-givers" became visible, both in Th e Moldovan population in Italy is predominantly made up of women (66%). It is also distinctive in that the average age is considerably higher than for immigrants of other origins: 23% are over 50 years old (for the immigrant population as a whole, this share is around 16%), while the percentage of minors (18%) is considerably low, compared with 24% for all regularly residing non-EU citizens (Ministero del lavoro e delle politiche sociali 2016). Over seven out of ten Moldovans live in Northern Italy, especially in the North-East (with a single region, Veneto, where 27% of the total are concentrated, and with signifi cant communities also in Lombardy and Piedmont). Next comes the Centre (about 21% of the total located in Emilia-Romagna and relevant presence in Lazio and Tuscany) and fi nally the South, with a marginal share of around 3% (almost totally in Campania) -see Figure 2.
Th e Moldovan population is also somewhat de-concentrated across many Italian cities. While many reside in Rome (more than 8,600), the majority of Moldovans live in middle-size provincial towns located in Veneto, Emilia Romagna and Lombardy: Parma (5,000), Venice, Padua, Bologna (over 4,000), Verona, Brescia, Modena, Reggio Emilia (3,500-1,500) are with the bigger cities of Turin (4,500) and Milan (3,000) in the top ten (ISTAT 2016).
Such geographic deconcentration is partly refl ective of a slow rise in the diversifi cation of employment sectors of Moldovans. Domestic assistance and personal services is still the dominant sector (53% of those working, see Graph 2) but other major sectors of economic activities are increasing, including industry and construction (18%), transportation and business services (12%), and wholesale and retail trade (11%). Crucially, domestic work has not been as deeply aff ected by the Italian economic crisis as other sectors and, consequently, the large number of women employed as caregivers may have mitigated the negative impacts on job levels for Moldovans: in 2015 the unemployment rate was 15% (about 20,000 jobless people).
Overall, 67% of the Moldovan population in Italy aged between 15 and 64 are employed. Th is is 9 percentage points higher than the rate recorded for non-EU citizens taken as a whole. Th is may refl ect the relative age composition of the population and the positively selected educational profi le (the majority of Moldovans employed in Italy have a medium to high educational qualifi cation, with 64% holding an upper secondary school certifi cate and 18% a university degree). However, that said, the prevalent area of Moldovan employment is unskilled and skilled manual labour, performed by 97% of Moldovans workers, compared to 3% of the total who are executives and professionals. Th e number of Moldovan owners of individual fi rms in Italy totalled 4,600 at the beginning of 2016: a low rate in absolute terms, but with a promising growth prospect if compared with the previous years.
The Moldovan community in Italy illustrates the diversification of migration strategies we introduced above. More than 65% of Moldovan migrants in the country hold a long-term residence permit. Th e increasing percentage of minors who have joined their parents abroad thanks to a residence permit for family reunifi cation and the growth of a second generation born in Italy have together driven the number of Moldovan students in Italian schools up to 25,000. One out of three of these Italian-Moldovan youngsters are enrolled in the upper secondary school, with 80% of them attending technical and vocational programs (Ministero del lavoro e delle politiche sociali 2016).

Czechia
Sustained Moldovan migration to Czechia is more recent than Moldovan migration to Italy, increasing only after 1989. Along with other streams of immigration to Czechia, Moldovan migration continued to grow until 2009 when the global recession increased the restrictions Czechia placed on immigration (ČSÚ 2016). As Moldovans in Czechia were considered "standard" economic migrants they were particularly impacted by restrictive  measures targeted at economically active foreign workers (Valenta, Drbohlav 2014). The overall number of Moldovans in Czechia has dropped by almost 50% from 2009 to 5,182 in 2016. There may be some signs this decline has now ended and numbers are stabilizing (see Graph 3). While the overall number of migrants is less than in Italy, there are important shifts in the gender composition of the Moldovan population. What had been an overwhelmingly male population had become, by 2016, more gender balanced, with women counting for 46% of all Moldovan employees, and 46.5% of all Moldovans in Czechia. As in Italy before 2002, this ratio may still be an undercount of women Moldovan workers in Czechia (see e.g. Ezzeddine et al. 2014) While lower than the ratio in Italy, it is higher than the ratio for other immigrant groups in Czechia, which average 36% female and 64% male. There are at least two possibilities for this trend. First, the continued feminisation of emigration from Moldova may have led to the identification of Czechia as a place of employment for care workers. Second, Moldovans may be pursuing a different integration strategy leading towards more family-based reunification possibly to access employment visas over the long term. In fact, the permanent residency (i.e. family-based migration) of Moldovans has been growing in both relative and absolute terms. The age composition is broadly similar to other foreign employed groups in Czechia. 50% of Moldovan employees are in the age group 31-45, 2.5% are younger than 20, and 16 per cent of them are older than 50 years (the corresponding figures for all foreign employees are 44.5%, 3%, and 18%, respectively).
In geographic terms Moldovans appear concentrated in central region of Czechia (Figure 3) with eastern and Moravian parts of Czechia home to fewer Moldovans. The district with the highest share of Moldovans compared to total population is Liberec (2‰) followed by Nymburk (1.9‰) and Beroun (1.8‰). In terms of the Moldovan population per se it is mostly concentrated in the biggest cities in Czechia. Almost 30% of the community is based in Prague, just over 6% in Liberec, 4.5% in Plzeň, and 3.8% in Brno and Nymburk each.
The structure of employment of Moldovans follows the overall figures on migrant employment in Czechia, notwithstanding any undercounting of particular groups as above. Major sectors of employment (NACE system) , are manufacturing, construction, and wholesale and retail trade (Table 1). The relatively high numbers of Moldovans (10.5%) in administrative and support service activities reflects some awkwardness of the classification -those counted in this sector are in fact employees of employment agencies and, thus, can in reality be employed in a variety of different industries. Furthermore, Moldovans predominantly work as employees in a company (Table 2). Crucially, the number of those listed as entrepreneurs has declined since 2011. As the global economic recession deepened, and Czechia sought to limit immigration, the number of Moldovans in Czechia in absolute terms declined significantly, while the share of unskilled partly decreased and of skilled workers actually increased slightly. In any case, the educational structure of employed Moldovans is less selective than foreign-born employees in Czechia. 42% of working Moldovans have a basic education compared to 30% of all foreign workers, Tab. 1 Major sectors of economic activity of Moldovans (employees) in Czechia, as of 31 July, 2016.

Sectors of employment (CZ -NACE)
Share of Moldovans on overall foreign employment while only 7.5% have a tertiary education compared to 17.7% of all foreign workers in 2016. The rather lower educational attainment of Moldovan workers in Czechia is further reflected by data on occupational status (ISCO classification is presented in Table 3). Moldovan employees show a considerable over-employment in categories with the lowest required qualifications.

Implications for Migration-Development Nexus
This section discusses how the preceding comparison of Moldovan migration in Italy and Czechia sheds light onto the concept of the migration-development nexus. In both countries Moldovans are a stable and important segment of the immigrant population. The population in Italy is larger (almost thirty times), longer established, more educated, more decentralized, and more feminised than the population in Czechia.
While Moldovan migrants to both countries are driven by socioeconomic considerations, of which livelihood remitting to family is key, Italy and Czechia have adopted very different migration policy regimes to manage mobility, and Moldovans have been impacted by this. In Italy a sort of "amnesty" in 2002 made the large population of women migrants visible. It may have been influential in leading today's high rate of family reunification and in the relatively resilient response to economic recession in 2009 when relatively few Moldovans left. By contrast, Czechia took a very efficient restrictive approach to temporary labour immigration following recession in 2009 which, in relative terms, decreased unskilled Moldovan migrants and, hence, increased skilled Moldovan migration slightly. In both countries, and beyond, Moldovan migrants have used increasingly diverse strategies to access work. This includes long-term or permanent residence permits, family reunification, and using Romanian identity.
We confirmed the expectation of the migration-development nexus (Faist 2004) that the Moldovan economy and, more widely, society depends upon migration and its financial remittances to combat socioeconomic problems. However, the extent to which migration policy is directly enabling development is less clear. We found that Moldovan international migration systems, and the strategies being used by migrants, are diversifying. Part of this diversification is geographic, with new destinations, including countries with no prior history of immigration, and those of the former Soviet Union. While migrants may choose destinations in accord with destination policy, in reality Moldovan cultural and linguistic affiliations affect patterns of migration. More widely, as remitting may be motivated by family considerations, not abstract macro-economic aspirations, it is planned over social horizons. This may be the reason for differential strategies of family reunification in Italy and Czechia.
These comparisons generate a series of research questions. For example, what are the mobility, employment, and transnational strategies of those Moldovans who have used family reunification in Italy and Czechia (Bailey, Boyle 2004)? Why has the Czech immigration policy led to a deskilling of Moldovan immigration? What are the long term prospects for development from migrant remittances? Without deep societal transformation changes in Moldova which must go hand in hand with improvement of the socioeconomic situation, emigration may continue with limited return to Moldova, with very limited positive effects upon developmentonly via remittances at an individual/family level and perhaps very limited and sporadic investments. In any case, no significant development would highly probably be visible at local, not to mention regional or nationwide levels (de Haas 2007b).

Summary and Conclusions
The general goal of our paper was to advance knowledge of contemporary Moldovan migration. With the deterioration of economic conditions since 1991, financial remitting has continued to be very significant, both to the overall economy and, increasingly, to many families for meeting basic livelihood needs. One in three Moldovan families are somehow affected by mobility as part of "an emigration culture". More women leave Moldova. Overall, Moldovan migration is diversifying, with new destinations, an expanded range of strategies for entering countries and accessing labour markets, and changing patterns of migrant selectivity, some seemingly affected by government policy. The global economic recession of 2008 had different implications for Moldovan migration patterns to and from Czechia and Italy.
Our specific objective was to assess these trends using the concept of the migration-development nexus. On the one hand, the concept provides a suitable explanation for why origin countries (Moldova), destination countries (Italy, Czechia) and migrants all perceive a "win-winwin" from financial remitting and, accordingly, orient their government policy and livelihood strategy respectively. However, as the case of Italy and Czechia make clear, social and political context matters, as Moldovans reacted very differently to global economic recession in these countries. To say that the nexus is a simple "pendulum" swinging back and forth is an unscientific and -in the currently charged political climate of migration debate -irresponsible characterisation of a diversifying and complicated migration system.
We thus end with a call for further research on the linked social and economic processes that make up the migration-development nexus, including social remittances (e.g. Levitt 1998 3 ). To address these and other questions we need robust data that permit comparative analysis of international migration. For example, in the Czech context, what is the extent of under-reporting of female migrants from Moldova and, indeed, other origins? Given the rapid diversification of contemporary migration, such data is urgently required. Because it must be harmonized cross-nationally an appropriate balance between securing personal data records and disclosing scientific information for the good of the whole society will need to struck.