ChapterPDF Available

Migration Experience of the Baltic Countries in the Context of Economic Crisis

Authors:

Abstract

The Baltic countries, which experienced intensive outflow of labor during the first five years after joining the EU provide an interesting case also for a study of migration response to economic shocks. The behavior of the Baltic migrants was different from that of their counterparts from other NMS. During the economic crisis of 2009-2010 and its aftermath, mobile citizens of other countries which joined EU in 2004, responded primarily to worsening economic situation in host old member states: emigration slowed down, while return migration intensified. By contrast, the behavior of the Baltic mobile citizens was, at large, driven rather by dramatic rise of unemployment and fall of household income in their home countries. New emigration wave emerged from each of the three Baltic countries: outflows doubled or almost doubled compared to the pre-crisis levels. More importantly, in 2013 emigration was well above the pre-crisis levels in all three countries despite resumed economic growth and massive outflow of population during the previous years. This chapter offers a theoretical framework and empirical evidence for understanding the patterns of emigration from the Baltic countries in the 21st century. The focus is on emigrants selectivity with respect to human capital, ethnicity and citizenship, as well as on labor market outcomes. We also assess the demographic and economic implications of recent emigration.
1
MIGRATION EXPERIENCE OF THE BALTIC COUNTRIES
IN THE CONTEXT OF ECONOMIC CRISIS
Mihails Hazans
University of Latvia and Institute for the Study of Labor (IZA), Bonn
Abstract
The Baltic countries, which experienced intensive outflow of labor during the first five years after
joining the EU provide an interesting case also for a study of migration response to economic
shocks. The behavior of the Baltic migrants was different from that of their counterparts from
other NMS. During the economic crisis of 2009-2010 and its aftermath, mobile citizens of other
countries which joined EU in 2004, responded primarily to worsening economic situation in host
old member states: emigration slowed down, while return migration intensified. By contrast, the
behavior of the Baltic mobile citizens was, at large, driven rather by dramatic rise of
unemployment and fall of household income in their home countries. New emigration wave
emerged from each of the three Baltic countries: outflows doubled or almost doubled compared to
the pre-crisis levels. More importantly, in 2013 emigration was well above the pre-crisis levels in
all three countries despite resumed economic growth and massive outflow of population during
the previous years.
This chapter offers a theoretical framework and empirical evidence for understanding the patterns
of emigration from the Baltic countries in the 21st century. The focus is on emigrants selectivity
with respect to human capital, ethnicity and citizenship, as well as on labor market outcomes. We
also assess the demographic and economic implications of recent emigration.
Keywords: Emigrant selectivity, human capital, brain drain, downskilling, return migration,
migrant network
JEL classification: J61, J24, J62, J15
Suggested citation:
Hazans, Mihails (2016). “Migration Experience of the Baltic Countries in the Context of
Economic Crisis”, in: Martin Kahanec and Klaus F. Zimmermann (eds), Labor Migration, EU
Enlargement, and the Great Recession. Berlin - Heidelberg: Springer (2016), 297-344.
DOI: 10.1007/978-3-662-45320-9_13
Financial support from European Social Fund Research Project 2013/0055/1DP/1.1.1.2.0/13/APIA/VIAA/040 “The
emigrant communities of Latvia: National identity, transnational relations, and diaspora politics“ (Institute of
Philosophy and Sociology, University of Latvia in cooperation with Faculty of Economics and Management,
University of Latvia) is gratefully acknowledged. The chapter greatly benefited from guidance and useful comments
by the editors, as well as from fruitful discussions with the authors of the other chapters in this volume during
IZA/CEUR Workshops in Budapest (2011-2012), and colleagues in other meetings. The usual disclaimer applies.
2
1. Introduction: The new exodus from the Baltics and its economic background
The Baltic countries, which experienced intensive outflow of labor during the first five years
after joining the EU (Hazans and Philips, 2010) provide an interesting case also for a study of
migration response to economic shocksi. The behavior of the Baltic migrants was different from
that of their counterparts from other NMS. During the economic crisis of 2009-2010 and its
aftermath, mobile citizens of other countries which joined EU in 2004, responded primarily to
worsening economic situation in host old member states: emigration slowed down, while return
migration intensified (Aujean (2012); European Commission (2012: Chapter 6, Chart 2); Zaiceva
and Zimmermann (2012: Figure 1); European Commission (2013: p. 43); Kaczmarczyk (2013:
Table 5.1)).
By contrast, the behavior of the Baltic mobile citizens was, at large, driven rather by dramatic
rise of unemployment and fall of household income in their home countries. To illustrate, Latvian
GDP dropped by almost 20% between 2007 and 2009 and the unemployment rate peaked in 2010
at almost 20%, a steep increase from 6% in 2007. Similar developments characterized Estonia and
Lithuania, with all three economies gradually recovering during the early 2010s.
As was predicted in Hazans and Philips (2010), new emigration wave emerged from each of
the three Baltic countries: outflows doubled or almost doubled compared to the pre-crisis levels in
2009 (Latvia), 2010 (Lithuania) and 2012 (Estonia), see Figure 5 below. More importantly, in
2012 emigration (both gross and net, in absolute numbers as well as in rates) was well above the
pre-crisis levels in all three countries despite resumed economic growth and massive outflow of
population during the previous years.
Another way to evaluate emigration flows from different countries is to look at the change in
stock of nationals of these countries in other EU/OECD member states. By this approach, Latvian
emigration stands out in the EU as the most dramatic response to the crisis. According to the 2011
EU Labor Force Survey (LFS), net inflow of Latvian nationals aged 15-64 to the rest of EU
between 2008 and 2011 was by half larger than between 2005 and 2008. A parallel increase for
Hungarian and Estonian citizens was found to be modest, while a decrease was recorded for
citizens of other CEE countries, see Aujean (2012). Moreover, Latvia is the only country among
the EU10 for which the estimated impact of recession on net migration to EU15 countries in
2008-2009 was positive (see European Commission, 2012, Chapter 6, Table 9, based on Holland
et al., 2011).
Overall mobility rate of nationals of given EU country can also be measured by the total (or
accumulated during some recent period) stock of working-age citizens living in another EU
member state, expressed as percentage of working age population of the country of origin.
According to estimates from EU LFS 2010, 2011 and 2013 (European Commission, 2012;
Aujean, 2012; Andor, 2014), among ten CEE member states Lithuania was ranked second in terms
of both total and recent mobility; Latvia was ranked fourth (respectively, third) on total
(respectively, recent) mobility by 2011 but moved to third (respectively, first) position by 2013.
Estonia was ranked sixth to seventh on all accounts. Being survey based, these figures likely
underestimate the size of Baltic diasporas (which constitute tiny proportions of population of most
of the receiving countries), but they support other evidence suggesting that Lithuanians and
Latvians have been recently among the most mobile CEE citizens. According to the most recent
3
estimates (Andor, 2014), economically active citizens of Lithuania (respectively, Latvia) which
have settled in other EU member states during the last 10 years accounted to 8.1% (respectively,
6.9%) of the origin country's labor force, well above corresponding figures for Poland (4.4%) and
Estonia (3.3%). Kahanec (2012) finds in Lithuania and Latvia the highest imminent migration
potential among EU member states.
The Baltic countries, despite being small in terms of population, also appear in the lists of top
origin countries in several European destinations: Lithuania was among top three origin countries
of foreigners in Ireland, Norway and Iceland and among top seven in the UK; Latvia was among
top five in Ireland and Iceland and among top 15 in the UK and Norway; and Estonia was the
number one origin of foreigners in Finland (OECD, 2014a: Table B.5).
From migration perspective, an important distinctive feature of the Baltic countries is a large
share of ethnic minorities (mostly Russian-speaking in Estonia and Latvia; Polish and Russian-
speaking in Lithuania); immediately before EU enlargement of 2004 they accounted for 41%,
32% and 16% of population in Latvia, Estonia and Lithuania, respectively. Moreover, at that time,
about one half of minority population aged 18-64 (note that nine out of ten emigrants depart at
this age) in Latvia and Estonia did not hold citizenship of these countries and thus were not
covered by the legal provisions for free movement of labor within EU; most of them had so-called
non-citizen passports of these countries, while others held Russian, Ukrainian or other citizenship;
see e.g. Tammaru and Kulu (2003), Hazans et al. (2008), Leping and Toomet (2008), Hazans
(2010, 2011a) for details.
This chapter aims at providing a detailed in-depth comparison of emigration patterns
experienced by the Baltic countries since the beginning of the 21st century and especially during
the post-enlargement decade (2004-2013). Section 2 compares economic and social context in the
Baltic countries and Poland. Section 3 discusses demographic challenges to the Baltic countries
caused by emigration. Destination-specific cross-country comparison of scale and trends of
emigration is provided in Section 4. This section also assess the deviations from reality found in
the emigration statistics of the sending countries. Section 5 offers a conceptual framework and a
set of hypotheses about the nature of the four emigration waves observed in 2000-2013.
Corresponding empirical analysis of the changes in emigrants profile (with a special focus on
selectivity of emigrants with respect to human capital and ethnicity), as well as in reasons for and
intended duration of migration, is found in Section 6, which also discusses differences between
emigrant composition acrosss destination countries. Labor market outcomes of emigrants are
discussed in Section 7. Section 8 focuses on return migrants. The economic impact of emigration
is discussed in Section 9. Section 10 concludes.
2. The economic and social context in Baltic countries and Poland, 2008-2013
Despite similar historic, economic and social context, there are also significant differences
between the three Baltic countries. Estonia, with the highest GDP per capita, the lowest (yet very
high) unemployment peak and the fastest decline in unemployment rate thereafter (see Figure 1),
entered the crisis and navigated through it in a better shape than its neighbors. Stabilisation fund
created in Estonia during the growth period was one of the factors behind this difference but also
an evidence for a better governance in general. On the other hand, both before and during the
crisis, Latvia featured the lowest GDP per capita and the highest and most persistent
4
unemployment rate of the three Baltic countries (Figure 1). Since 2011, unemployment was
falling in all three Baltic countries, remaining at two-digit levels though in Latvia and Lithuania.
By comparison, in Poland economic growth continued in 2009-2013 (though at a smaller
pace than before), while unemployment stayed at about 8% in 2009 and at 10% in 2010-2013;
GDP per capita was higher than in Latvia but lower than in Estoniaii.
Generally, Latvian population went through more difficult times than their Estonian,
Lithuanian or Polish counterparts. According to EU-SILC data, 18% to 22% of Latvian
households faced great difficulty making ends meet in 2009-2012, while this share varied in the
range of 11% to 13% in Lithuania, 8% to 9% in Estonia and 12% to 14% in Poland. During the
same period, 22% to 26% of Latvian households had arrears on mortgage or rent payments, utility
bills or hire purchase, compared to 10% to 13% in Lithuania and Estonia and 14% to 15% in
Poland. Importantly, incidence of mortgage or rent arrears was below 2% in Lithuania an Poland,
below 3% in Estonia, but between 5% and 7% in Latvia.
Figure 1 Real GDP per capita and unemployment rates, the Baltics and Poland, 2005-2013
Source: Eurostat
To survive the crisis, Latvia was forced to apply for emergency financial assistance from the
EU, IMF and the World Bank, while Estonia and Lithuania managed without external help and
experienced much more modest wage cuts than Latvia (European Commission 2011, Graph I.3.1).
Moreover, in Latvia the Great Recession has been perceived by a majority of population as a
systemic (rather than just a financial) crisis, which was less pronounced in Lithuania and was not
the case in Estonia. Figure 2 illustrates the differences in popular perceptions between the Baltic
countries and provides comparison with Poland.
0
5
10
15
20
25
5,000
5,500
6,000
6,500
7,000
7,500
8,000
8,500
9,000
9,500
10,000
2005
2006
2007
2008
2009
2010
2011
2012
2013
Unemployment rate, %
Real GDP per capi ta, EUR
Estonia unemp.
Lithuania une mp.
Latvia unemp.
Estonia GDP
Lithuania GDP
Latvia GDP
Poland GDP
5
Figure 2 Satisfaction and trust indicators in the Baltic countries and Poland, 2008-2013
Notes: Satisfaction and trust are measured at the 0-10 scale. The Figure reports mean values
(excluding non-response). Standard errors are between 0.04 and 0.06 in all cases but Lithuania 2011 (0.07 to
0.09). Source: Calculation with the data of European Social Survey.
As shown in Blanchflower and Shadforth (2009), life satisfaction seems to be an important
determinant of migration; hence, one can infer from Figure 2 something about future migration
patterns. Before the crisis, Estonia featured the best satisfaction measures on all accounts; Latvia
and Poland shared similar values for all indicators except those related to economy and education,
which were higher in Poland.
During the crisis, satisfaction with the state of economy, the national government and the way
democracy works, as well as trust in parliament, fell dramatically in Estonia and especially Latvia
(the Lithuanian satisfaction levels were similar to the Latvian ones, while pre-crisis data are not
available for Lithuania), but increased in Poland. Nevertheless, Estonia remained on top in terms
of trust in parliament, as well as satisfaction with education and health systems, and shared with
Poland the highest "mark" received by the government (this mark was just 3.5 on the 0-10 scale,
though). Poland ranked first on the state of economy and on democracy. Among the four
countries, Latvia featured the lowest indicators on all accounts, and Lithuania was slightly above
(except for education, where this order was reversed), while Poland and Estonia were doing much
better.
Results of two post-crisis surveys (conducted between 2010q4 and 2013q2) are available for
all countries but Latvia. By 2010, all indicators for Poland (except the one on health) and Estonia
were above the levels observed in the crisis years. However, in Estonia, trust in the parliament and
satisfaction with the economy and the government were still below the pre-crisis level - and fell
even further by 2012. In turn, all Polish indicators by 2012 fell to the levels similar to those
observed in 2008-2009 or slightly lower.
Lithuanian popular sentiment indicators, starting from extremely low levels during the crisis,
went up both between 2009 and 2011 and especially between 2011 and 2013.
0
1
2
3
4
5
6
7
2006q4
2008q4-2009q1
2010q4
2012q4
2006q4-2007q1
2008q4-2009q1
2010q4
2012q4
2007q2q3
2009q2q3
2009q4
2011q2
2013q2
Polan d
Estonia
Latv ia
Lithuania
Ave rage satisf action /tr ust (m ax=1 0)
the state of e con omy
the national government
trust in parliament
the way democ rac y wo rks
the state of e duc ation
the state of health services
6
When comparing situation in Poland, Estonia and Lithuania at the end of 2012 or beginning
of 2013, one finds that the economies of the three countries received almost identical marks (4 at
the 0-10 scale); at about the same level were the best of the three assessments of the government
(4.3 for Lithuania, followed by Estonia with 3.9 and Poland with 3.4) and the parliament (3.9 for
Estonia, 3.1 for Lithuania and 3.0 for Poland). The remaining indicators were somewhat higher:
about 5 for democracy in all three countries, between 5 and 6 for education and between 3.5 and 5
for health services, Estonia being on top in the latter two cases.
To sum up, the above discussion of economic and social context suggests that during (and
hence, due to the network effect, after) the crisis, emigration could be a more popular coping
strategy in Latvia and Lithuania than in Estonia (and than in Poland, too), while more intensive
return migration is to be expected in Poland and Estoniaiii. Furthermore, the dynamics of push
factors suggests that in the post-crisis period emigration could increase in Poland and Estonia but
decrease in Lithuania. Moreover, it seems plausible that the satisfaction levels like the ones
observed in Latvia and Lithuania in 2009 (as opposed to those found in Estonia and Poland) signal
a steep rise in emigration rates which are likely to remain high for quite some time after
satisfaction rates rebounce.
3. Baltic populations: "Natural" shrinking and aging reinforced by emigration
Since the beginning of the 21st century, loss of population due to emigration reinforced
negative natural change in all three Baltic countries. In 14 years (2000-2013), Latvia's and
Lithuania's populations declined by about 17% and Estonia's - by 9% (Figure 3). Natural change
accounted for more than one-third of this loss in Latvia, about one-third in Estonia and slightly
less than one-fourth in Lithuania. During the economic crisis and its aftermath overall
depopulation intensified in all three countries, but the share of natural decline in the total change
was much smaller than before (Figure 3).
7
Figure 3. Natural decrease of population and net emigration.
The Baltic countries, 2000-2012.
Notes: The 20002003 category covers 4 years, while 20042008 and 20092013 cover 5 years each. In all
three Baltic countries, annual average net emigration rates (as implied by the Figure) in 20092013 are much higher
than in the previous periods. Source: Eurostat, OECD and own calculation.
Most of Baltic emigrants depart at age between 15 and 34 years, hence this age group shrinks
faster than population in general, thus accelerating aging (caused also by declining birth rates) and
putting at risk sustainability of social security system.
Figure 4 illustrates the effect of post-enlargement migration on youth cohorts by tracking
their size (which is almost unaffected by natural change) over the period of 2004-2013. In ten
years, Latvian cohorts aged 15 to 19, 20 to 24 and 25 to 29 years at the beginning of 2004, have
sent abroad, respectively, 20%, 19% and 15% of their members, while corresponding Lithuanian
cohorts in the same period lost to migration 27%, 26% and 17%. The crisis and post-crisis five
years (2009-2013) account to most of these human losses. Note that these data are based on the
official population statistics, which, especially in Latvia, underestimate emigration in the post-
crisis period (Hazans 2013c), so actual losses are likely even larger.
Figure 4. Change in the size of selected age cohorts.
Latvia and Lithuania, 2004-2013 (in % of population on January 1, 2004)
10.8% 13.5%
6.1%
0%
2%
4%
6%
8%
10%
12%
14%
16%
18%
2000-2003
2004-2008
2009-2013
2000-2013
2000-2003
2004-2008
2009-2013
2000-2013
2000-2003
2004-2008
2009-2013
2000-2013
Latvia
Lithuania
Estonia
% of population on 01.01.2000
Net emigration
Natural decrease
8
Source: Statistics Latvia, Statistics Lithuania and own calculation
Data presented in Figures 3-4 highlight serious demographic problems faced by the Baltic
countries. To put these in international perspective, let us compare crude birth rate and crude rate
of natural change of population observed in the Baltic countries with those found in other Eastern
NMS, as well as in main destination countries of the Baltic emigrants (the UK, Ireland, Germany,
Norway and Finland). It appears that the "natural" aging caused by declining birth rates is much
more pronounced in the Eastern EU member states than in the comparison countries. Furthermore,
in terms of either recent birth rates or post-enlargement rates of natural change of population,
Latvia and Lithuania are among the three "demographically worst cases" in the Eastern part of
EU, while Estonia performs significantly better; average birth rate in 2009-2012 was lower than in
2004-2008 in Latvia, while Estonia and Lithuania have seen an increase. Finally, all three Baltic
countries feature substantially lower birth rates and rates of natural change than the UK, Ireland
and Norway; moreover, in terms of the rate of natural change Latvia and Lithuania perform worse
than Germany, and Estonia performs worse than Finland. In other words, the countries which host
most of the Baltic emigrants have better demographic prospects than the Baltic countries.
To sum up, after EU enlargement in 2004, and especially during the crisis and post-crisis
years of 2009-2012, emigration from Baltic countries reached levels that threaten reproduction of
their populations, sustainability of social security systems and economic development. The
demographic risk is most pronounced in Latvia. According to a survey conducted in 2012, three
quarters of Latvia's population perceive emigration as the single largest threat to the country and
its people (Hazans 2013c, Figure 4.2). Noteworthy, families with the largest demographic
potential (the ones with children or planning to have a child within three years) are more likely to
-30%
-25%
-20%
-15%
-10%
-5%
0%
Latvia
Lithuania
Latvia
Lithuania
Latvia
Lithuania
Latvia
Lithuania
10-14 years
15-19 years
20-24 years
25-29 years
Age on 01.01.2004
2009-2013
2004-2008
9
emigrate (Hazans, 2013d; 2014a; 2014b). This finding is supported by the UK Population Census
2011 data, according to which total fertility rate of females born in Latvia and Lithuania was 2.51
and 2.29, respectively, well above corresponding rates observed at the same time in the home
countries (1.33 and 1.55, respectively).
4 Baltic emigration during and after the economic crisis: the scale, the trends and the
geography
In this section we use gross outflows from the Baltic countries to analyse the trends in
emigration and to assess the deviations from reality found in the emigration statistics of the
sending countries.
While economic crisis started in the second part of 2008, its full effect on migration can hardly
be found in data before 2009. Taking year 2008 data as the "pre-crisis benchmark", Figure 5
presents gross outflows of nationals from the Baltic countries (broken down by main destinations
in EU and OECD) in 2008-2013. The outflows are measured per 1000 population of the sending
country at the beginning of 2008 (this approach ensures comparability of the outflows across
countries in the relative terms, as well as across time in absolute terms). The underlying data are
obtained by putting together Eurostat and OECD data on immigration of foreigners by nationality;
for Ireland and the UK data on allocation of social security numbers (PPSNs and NINOsiv,
respectively) are used instead, because British and Irish immigration data are survey-based and
severely underestimate inflows from such small countries as the Baltic ones.
In 2008, total gross outflow of nationals from Lithuania (respectively, Latvia and Estonia)
accounted to 1.2% (respectively, 0.9% and 0.6%) of population on 01.01.2008.
During the five crisis and post-crisis years (2009-2013) gross outflows from Estonia accounted
for about 4.5% of their population, while outflows from Latvia and Lithuania were equivalent to
9.1% and 9.6% of their populations (recall that we stick to population as of 01.01.2008). This is
consistent with our expectations (based on the analysis of the economic and social context in
Section 2 above) that during and after the crisis emigration could be a more popular coping
strategy in Latvia and Lithuania than in Estonia.
In comparison with the year 2008, average annual emigration rate in 2009-2013 more than
doubled in Latvia, increased by by more than a half in Lithuania and by 40% in Estonia.
Figure 5 Gross emigration of nationals from the Baltic countries
to main EU and OECD destinations, 2008-2013
10
Sources: Eurostat and OECD data on immigration of foreigners by nationality; Ireland and the UK data on allocation
of social security numbers. For 2013, data were available on outflows to the UK, Germany, Nordic countries, Ireland,
the Netherlands and Austria (these destinations covered more than 90% of outflows in 2012); the remaining flows
were (conservatively) predicted.
Emigration dynamics during the crisis also varied across the Baltic countries. Latvian and
Lithuanian emigration feature a skewed inverse U-shaped pattern: a steep increase in 2009-2010
followed by a less steep and smaller decline (in 2011-2012 for Latvia and in 2012 for Lithuania);
this pattern is found also in flows to the UK and to the Nordic countries; flows to Germany (as
well as much smaller flows to BENELUX countries, Austria and Switzerland) were steadily
growing at least until 2011 but experienced modest declines in 2012 and/or 2013.
Total gross emigration of Estonian nationals was growing in 2009-2012 and stabilized in
2013; the growth was driven mainly by ouflows to Finland, the UK and Germany; the latter two,
however, reached their peak values in 2011.
In 2012, total outflow of nationals from Latvia (respectively, Lithuania and Estonia) to the EU
and OECD destinations exceeded the 2008 level by 77% (respectively, 49% and 52%). While
outflows from the Baltic countries to Ireland declined by more than a half each, a substantial
increase is found in the outflows to all other main destinations: outflows from Latvia and
Lithuania to the UK increased by more than two-thirds; total outflow of Estonian (respectively,
Latvian and Lithuanian) nationals to the Nordic countries increased by more than 80%
(respectively, almost 80%; 60%), while outflow to Germany from Estonia (respectively, Latvia;
Lithuania) more than doubled (respectively, more than quadrupled; almost tripled).
Plausibly, the emigration-boosting effect of the crisis in the Baltic countries included both
transitory and persistent components. The latter can be explained by combination of several
forces, including the network effect, non-economic push factors (such as disappointment and loss
of perspective) and insufficient labor demand in the home countries.
0
5
10
15
20
25
2008
2009
2010
2011
2012
2013
2008
2009
2010
2011
2012
2013
2008
2009
2010
2011
2012
2013
Estoni a
Latv ia
Lithuania
gross outflows per 1000 of population at 01.01.2008
OECD non-Europe
Austria, Switzerland, Fr ance, Belgium,
Netherlands, Luxemburg
Italy, Spai n, Greece , Por tugal
Ireland
Nordic countries
Ger many
UK
All destinations -sending c ountries data
11
In Latvia and Lithuania, outflows to the UK and Germany accounted for about three quarters
of the total gross outflow to the EU and OECD countries in 2012; when Nordic countries and
Ireland are added, this share reaches nearly 90% in each of the three Baltic countries.
Finally, in 2013, emigration from each of the Baltic countries was just slightly below the (post-
crisis) 2012 level.v
Comparison of receiving and sending countries' data (also available from Figure 5) suggests
that in all three Baltic countries alike, the official data underestimate the scale of emigrationvi and
distorts its dynamics; see Hazans (2013c: Annex A4.A1) for a more detailed discussion.
5 Four emigration waves in a dozen years: the changing characteristics of Baltic EU
mobility
Recent history (in 2000-2013) of emigration from the Baltic countries can be loosely divided
into four episodes: (i) Pre-accession period (which we denote as 2000-2003); (ii) Post-accession
period of economic growth, to which we refer as 2004-2008 (although the crisis hit Latvia and
Estonia already at the end of 2008 its effect on emigration first appears only in 2009); (iii) The
crisis period: 2009-2010; (iv) the post-crisis period: 2011-2013. During this relatively short time,
the main reasons for emigration, emigration rates, the most popular destinations, as well as the
profile of the emigrant population and emigrants' plans, have changed substantially several times.
According to the human capital model of migration decisions (Sjaastad, 1962; Borjas, 1987,
1999) an individual (or a family) decides to move if expected (over the planning period) utility in
the host country (net of total cost of migration) exceeds utility in the home country. The
“calculation” should account for all factors which can affect the quality of life, including job
finding and job losing probabilities, expected earnings, legal status, career perspectives, working
and living conditions, generosity of social security system, social and cultural norms, perceived
life perspective for children, etc. The costs of migration, in turn, include monetary and effort costs
related to acquiring necessary information, job search, transportation, and maintaining the
connections with the country of origin, as well as psychological costs related to missing people
and environment left behind, uncertainty associated with the life in the new country, and
adaptation to the new reality. This framework helps to understand the patterns of selectivity of
emigrants and the way these patterns change over time in response to economic, political and
social developments in the source countries and in the potential host countries.
The pre-accession wave: personal characteristics
Before joining the EU (in 2000-2003), the Baltic countries featured two-digit unemployment,
while GDP per capita (at PPP) was well below 50% of the EU-15 average. Earnings of an
unskilled worker in the United Kingdom, Germany or the Nordic countries looked very attractive
in comparison with average earnings in the Baltics.vii These strong push and pull factors resulted
in a sizeable emigration potential (see Rose, 2000: 34 and Hazans, 2012: Figure 6.2 for details).
This potential was larger among ((Russian-speaking)) minority population, which, in comparison
to natives, featured less favorable (on average) labor market outcomes (Figure 6 below; see also
Hazans, 2010; 2011a) on one hand, and a weaker attachment to the home country, on the other
12
(see e.g. Rose, 2000: 64-66; Aptekar, 2009; Zepa and Kļave, 2011: Boxes 1.3, 1.9, 1.16, 1.20,
1.21; Anniste and Tammaru, 2014).
However, actual emigration rates in early 2000s weren't high, not least because of an
institutional environment which was not favorable to economic migration (need for work and
residence permits), but also due to very high migration costs (high transportation and
communication costs, limited availability of good quality internet, absence of convenient
extensive sources of information on vacancies and living and working conditions abroad). During
a four year period before accession (2000-2003), net outflow from Estonia (respectively, Latvia
and Lithuania) was about 1% (respectively, 1.5%; 2%) of population, see Figure 3.
To understand who were the likely movers in the pre-accession period, one should notice
that migration costs would have been relatively lower for persons with professional or at least
private contacts in potential destinations, good foreign language and IT skills, and opportunities to
use the internet for private purposes at the workplace. Clearly, all these attributes are more often
found among university graduates. On the other hand, absence of a favorable legal framework,
restricted access to reliable information, difficulties in job search "from overseas", as well as a
high risk of fraud by domestic firms recruiting workers for jobs abroad in early 2000s, implied
that emigration required a high degree of initiative and willingness to accept risk; these qualities
could be substituted by access to migration networks related to previous waves of migration
to/from the United States, Canada, Australia, Sweden and Germany, as well as Russia, Ukraine
and Belarus. Most emigrants driven by own initiative (rather than networks) were oriented
towards relatively new directions, mainly the United Kingdom and Ireland, where language
barrier for them was lower than in the rest of the EU, while migration costs were lower than to
other English-speaking countries. The pre-accession wave of emigration thus featured a
substantial positive selectivity on human capital and other personal characteristics, over-
representation of Russian-speakers, as well as a high degree of geographical diversification.
Figure 6. Ethnic gaps in employment and unemployment in the Baltic countries, 2000-2013
13
Notes: The gaps are defined as the differences between employment (respectively, unemployment) rates of
native and minority (respectively, minority and native) population aged 15-64. Sources: Statistics Estonia online
database; calculation with Latvian LFS data (2002-2013), Lithuanian LFS data (2002-2003) and ESS data (2009,
2011).
Post-accession emigration: Institutional and market factors
During the first five years within the EU (before the effect of the Great Recession on
migration patterns became apparent) migration flows in the Baltics were shaped mainly by
institutional and market factors.
Gradual implementation of free movement of labor within the EU (see Kahaanec et al (2014:
Table 1) substantially lowered both monetary and non-monetary costs of job search abroad and
migration, as well as the human capital threshold (in terms of skills, initiativeness and risk taking)
for labor migration. Kahanec et al (2014) use panel of intra-EU migration flows to show that
enlargement as well as removal of transititonal arrangements lead to more migration.
Together with high and growing demand for migrants’ labor in the EU15, this triggered a
sharp and persistent increase in emigration rates (see Figure 5; see also Kahaanec et al, 2014).
This, in turn, further lowered migration costs via migrant networks, rich social and media
infrastructure within rapidly growing Baltic diasporas in Ireland, the United Kingdom, Sweden,
Germany and elsewhere in old member states (see Hazans and Philips, 2010; OECD, 2012) and
the scale effect, which caused air and land transportation costs, as well as international phone
calls tariffs to fall; communication costs have been also pulled down by increased coverage and
speed of internet connections.
In addition, sufficiently strong pull factors, such as higher income and better working
conditions abroad (mentioned, respectively, by 60% to 70% and about 50% of potential movers
from the Baltic countries in 2005), as well as family- or friends-related factors (mentioned by 13%
0
2
4
6
8
10
12
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
% points
Estonia ethnic em ployment
gap
Latvia ethnic employm ent
gap
Lithuania ethnic
employm ent gap
Estonia ethnic
unemploym ent gap
Latvia ethnic unemploym ent
gap
14
to 22%) were at work; together, these factors covered about 80% to 90% of potential emigrants
from each of the three countriesviii.
On the other hand, due to strong economic growth in the Baltic countries, the unemployment
rates there were falling while real income was rising (see Figure 1 above; see also Hazans and
Philips 2010: Section 7 and Figure 12), gradually reducing expected gains from emigration. Thus,
during the second part of the post-accession period, motivation to move abroad driven by push
factors was falling, whilst motivation to return among recent emigrants was on the rise.
As the net result of the developments briefly described above, in the five post-accession years
Estonia, Latvia and Lithuania lost to emigration, respectively, 2.0%, 3.2% and 5.3% of their
population (see Figure 3).
The choice of destination countries during the post-accession period was of course srongly
affected by institutional factors: since May 1, 2004, emigration flows from the Baltic countries
became heavily oriented towards the UK, Ireland and (to a smaller extent) Sweden, following
decisions of these countries to open their labor markets for workers from the NMS. Likewise, a
sharp and persistent increase of the Finland's share in Estonian emigration is observed since 2006.
In what respects were the post-accession emigrants different from the pre-accession ones?
First, in 2004–2008 emigrants’ self-selection in terms of human capital was driven not so much by
individual’s comparative advantage in lowering migration costs, but mainly by expected gains in
terms of income and working conditions. These gains were, on average, larger for persons with
secondary or lower education. For instance, in 2005, tertiary educated employees in Latvia earned
by 54% (respectively, 76%) more than otherwise similar workers with secondary (respectively,
less than secondary) education (Hazans, 2007: p.18 and Figure 2.1). By contrast, in the EU-15
countries returns to schooling for post-acession immigrants from the Baltic countries and other
NMS were quite low, not least because majority of tertiary-educated members of this group held
jobs which did not require higher educationix.
Hence, one should expect that, in comparison with the pre-accession period, post-accession
emigrants from the Baltic countries are, as a group, less educated - either in absolute
(composition) or in relative (selection) sense, or both.
The effect of ethnicity and citizenship on propensity to emigrate has also changed. Due to
strong economic growth and labor shortage caused by emigration (see e.g. Hazans and Philips
2010: Section 7 and Figure 12), as well as gradual improvement in state language skills among
young and middle-age minorities (Hazans 2010: Figure 3; Hazans 2011a: Tables 8.8-8.9), labor
market position of ethnic minorities in 2004-2007 was steadily improving in Estonia and Latvia
(see Figure 6) and, plausibly, also in Lithuania. On the other hand, a substantial part of minority
population – those without Estonian or Latvian citizenship was not covered by the legal
provisions for free movement of labor within EU. Indirectly via spouses who held Estonian or
Latvian citizenship, as well as via migrant networks new migration possibilities emerged also
for non-citizens; yet their mobility opportunities in comparison to citizens worsened.
The above considerations suggest that, in comparison with the pre-accession period, post-
accession emigrants from the Baltic countries feature a significantly lower proportion of ethnic
minorities, especially non-citizens.
15
Another important feature of this emigration wave (which could not be predicted based on
theoretical considerations alone) is its mixed nature: while migration was to a large extent short-
term and/or cyclical ( see e.g. Hazans and Philips, 2010, Section 6, Figures 9 and 10), the Baltic
diasporas abroad were steadily growing.
Crisis-driven emigration (2009-2010): Lost jobs, lost perspectives, "new movers"and shift
towards permanent emigration
During the years of the Great Recession (2009-2010), push factors (mainly joblessness and
wage cuts, but also implied inability to pay back creditsx), were at work again. The psychological
shock was no less painful: a large proportion of people of working age (including those who
managed to keep their jobs) lost confidence in the future (see Hazans 2011b, 2013c; Saukienė
2011); as discussed in Section 2 this was most pronounced in Latvia, and least pronounced in
Estonia.
Finding a job in Western Europe was not as easy as before the crisis (hence, the role of
diasporas and informal networks increased). Yet it was much easier than in the Baltics. The rate of
unemployment was very low (3% to 4%) in Norway, the Netherlands and Austria, and remained
modest (about 8%) in the United Kingdom, Germany, Sweden, Denmark and Finland (European
Commission, 2010, Table 24). During 2009-2010, the job vacancy rate (i.e. the number of
vacancies relative to the sum of vacancies and occupied posts) in these countries (excl. Sweden)
was five to ten times higher than in Latvia, three to more than four times higher than in Lithuania
and about twice as high as in Estonia (European Commission, 2010, Chart 6). Lifting restrictions
on free movement of workers from EU8 countries by Belgium, Denmark and especially Norway
since May 2009 further facilitated labor migration to these destinations.
Moreover, nominal earnings continued to rise across old member states, while real earnings
did not decline (European Commission, 2011, graphs I.1.8, III.A3.5). Thus, expected gains from
emigration in terms of employment and earnings increased in comparison to the pre-crisis period.
In addition, as long-term joblessness was becoming more widespread in the Baltic countries,
the issue of social protection, previously neglected by the middle class, has gained importance as
a factor driving the migration decisions. Note that the Baltic countries feature very low income
replacement rate by unemployment benefit for long-term unemployedxi; for Latvia and Estonia,
this is the case also when social assistance and housing benefits are accounted for (European
Commission, 2011, graphs II.2.3- II.2.4); moreover, child benefits in the Baltic countries
(especially Latvia and Lithuania) were exteremly low in comparison with those paid in the main
destination countries for the Baltic emigrants.
High and persistent unemployment, weak social security system, lost perspectives these
were the factors that converged to make emigration a real option in the minds of Baltic residents,
even those who had not considered such a possibility before. There were two kinds of these "new
movers" : i) individuals who are inherently not very mobile but did not see another way out of
trouble; and ii) persons who were not satisfied with the developments in the home country and
with their own prospects there (even if they did not experience immediate economic hardship).
16
Unlike the pre-accession emigrants, most of those who left during (and after) the crisis were
not risk-takers: on the contrary, they perceived staying as too risky, and destination countries as
safe heaven.
Naturally, this implied a strong shift from temporary emigration of breadwinners towards
long-term or permanent emigration of entire families (see Table 1 for evidence based on the daily
records of EURES consultants in Latvia).
Table 1. Changes in the profile of EURES clients in Latvia, 2004-2010
2004-2007
2008-2010
Planning to move alone
Planning to move with family
Looking for temporary, low-skilled job
Looking for permanent, skilled job
Minimal knowledge of foreign languages
Better knowledge of foreign languages, higher
qualifications
Planning to return
Interested in legal employment and social protection
Source: Hazans (2013c: Table 4.6)
Evidence from survey of Latvian emigrants conducted in 2014 and surveys on emigration
intentions in Estonia conducted in 2006, 2010 and 2013 presented in Figure 7 confirms that during
the crisis years importance of both economic and non-economic push factors, betters social
security abroad, as well as family-related factors sharply increased in both countries.
How and why did the crisis change the emigrants' profile? In all three Baltic countries,
joblessness increased particularly among individuals without higher education (and even more so
among those who have completed only basic school or less)xii; On the other hand, in Latvia and
Estonia, relative labor market position of ethnic minorities (especially Latvian and Estonian non-
citizens) deteriorated during the crisis (see Figure 6 above and Hazans (2010: Figure 9; 2013c:
Table 4.5); in Latvia, it was accompanied by strengthening of the state language proficiency
requirements in the private sector (Hazans 2010:151; 2011:187).
Hence, based on (domestic) economic factors alone, one should expect a significant increase
in the proportions of the low-skilled and (in case of Estonia and Latvia) of the Russian-speakers
among emigrants (see Hughes (2005) and Ivlevs (2013) for some theoretical considerations, and
Hazans (2013c) for intentions-based empirical evidence), whilst there is no reason to believe that
brain drain will intensify. On the other hand, in times of crisis, the low-skilled might find it
difficult to compete with secondary school graduates for jobs abroad (one of the reasons being
poor language skills). Moreover, given that in Latvia and (to some extent) Lithuania the crisis was
perceived as systemic, the proportion of high-skilled among Latvian and Lithuanian emigrants
could also rise, because people who have opted to invest in higher education are usually future-
oriented.
17
Figure 7. Prevalence of various reasons for emigration among emigrants from Latvia
(2000-2014, upper panel) and potential emigrants from Estonia (2006-2013, lower panel)
Sources: Latvia (upper panel) - calculations with emigrants' survey data. Estonia (lower panel): Emigration
intentions surveys' data reported in Tarum (2014) and own compilation.
The ethnic story is also not straightforward, as a number of factors work against expected
shift towards higher proportion of minorities among emigrants. First, Estonian and Latvian non-
citizens (as well as residents holding citizenship of Russia and other CIS countries) are not
covered by the free mobility provisions. Second, there is anecdotal and media evidence that
emigration of young ethnic Latvians, especially those coming from small towns and countryside,
20%
30%
40%
50%
60%
70%
push
pull
push
pull
pull
push
pull
push
pull
pull
Economic and
social sec urity
factors
Political
factors, future
prospects,
quality of
state, quality
of life
New
exp.
Family
Economic and
social sec urity
factors
Political
factors, future
prospects,
quality of
state, quality
of life
New
exp.
Family
Men
Women
2000-2003
2004-2008
2009-2011
2012-2014
0%
20%
40%
60%
80%
Unemployment in
Estonia
Poor employment
prospects in Estonia
Higher earnings
abroad
Be tter s ocial sec uri ty
system abroad
Disappointment
with life in Estonia
Employees are more
respec ted abroad
Better quality of life
abroad
Desire for
professional
development
Be tter wo rk
organization abroad
Desire to br oade n
horizons, get new
experience
Per sonal or family
Push
Pull
Push
Pull
Pull
Economic and social security factors
Quality of state, quality of life, quality of job
Other
Other
2006
2010
2013
18
is growing because the share of those able to communicate in Russian is falling, and without
Russian language skills it is difficult to find a job in big cities. Finally, for ethnic Estonians,
emigration to Finland (which hosts most of Estonian emigrants) is easier than for non-Estonians
because Estonian language is similar to Finnish (Praakli, 2011).
The post-crisis wave (2011-2014): Emigration as "the new normal".
In the aftermath of the Great Recession, despite economic recovery, there have beeen no
clear signs of a considerable slowdown in emigration from the Baltics; moreover, potential for
further emigration is high and growing. Pull factors gained importance among the drivers of
emigration; furthermore, while economic reasons for emigration remain widespread, non-
economic ones become increasingly important, especially among university-educated population
(see Figure 7 above; see also Hazans, 2011b; 2013c; Saukienė, 2011; Samoškaitė, 2012). In
terms of destinations, Germany, which opened its labor market for EU-10 workers in 2011,
increased its share in Baltic emigration flows (see Figure 5); plausibly, this had an impact also on
composition of these flows, as Germany is more attractive than, say, the UK, for middle-aged
skilled manual workers.
Vast majority of Baltic populations now have close relatives or friends who have moved
abroad. Migration flows are shaped by these migrant networks, along with already formed but not
yet implemented emigration intentions (Hazans, 2013c; Veidemann 2010; Tarum 2014); in 2012-
2013, 8% of adult population in Estonia and 12% in Lithuania planned to move abroad
permanently (Hazans 2015b:16). Emigration has become “the new normal” (Hazans, 2014b), and
the Baltic diasporas will keep growing in the years to come.
6 The changing selectivity of emigrants: 2000-2014
Human capital
Analysis of trends in the "brain drain", i.e. the patterns of emigrants' selectivity on human
capital is complicated by a rather strong positive trend in skill composition of the Baltic
populations during the whole period of 2000-2013. To facilitate comparison across time, we use
selectivity index SI = ln(GM/GS) , where GM and GS are shares of university graduates (or any
other group of interest) among movers (i.e., emigrants) and stayers, respectively; thus, SI is
positive (negative) if tertiary educated persons are over-represented (under-represented) among
movers (Hazans, 2011b; 2012; 2013cxiii).
We start by using national LFS data on Baltic workers abroad reported as household
members in the home country (the analysis here extends that in Hazans 2012: 183-187).
According to LFS rules, these are "recent" emigrants, away from home for less than a year, which
allows treating data as "flow". Data presented in Figure 8 (top panel) support, for each of the three
countries, the hypotheses put forward in the previous section: university graduates were over-
represented among pre-accession emigrants (this was least pronounced in Estonia); both the share
of tertiary-educated among emigrants and the correponding selectivity index "at departure" (which
compares, for each period, emigrants with sending country's population aged 18-64 in the same
period) drop in the post-acession periodxiv (reflecting higher expected gains for the low- and
medium-skilled) and, except for the case of Estoniaxv, take higher values during and after the
19
crisis (reflecting rise of general disappointment and non-economic reasons for emigration in
Latvia and Lithuania).
On the other hand, for Estonia and Latvia, both the share of low-skilled among emigrants and
corresponding selectivity index is higher in the post-accession period than before (supporting the
idea that free movement of labor lowered human capital threshold for migration) and further
increases during and after the crisis (reflecting the fact that the low-skilled suffered stronger and
longer from the recession-related joblessness). For Lithuania, the share and selectivity index of
low-skilled slightly decreased after EU enlargement, but less so than respective indicators on the
tertiary-educated, so the Lithuanian data are also consistent with the notion of post-accession
emigrants being less educated. During the crisis, Lithuanian low-skilled (in contrast with their
Estonian and Latvian counterparts) were even stronger under-represented among (recent and
having family left behind) emigrants (probably they, due to poor language skills, were less
inclined to emigrate when job finding in the destination countries became more difficult).
We proceed by using Population Census (or Population Register) data on educational
attainment of adult (aged 15+) Baltic-born residents of European OECD countries early in 2011,
depending on the arrival period. These are stock data, and emigrants' education could be
completed also after leaving the home country. Therefore we use "age-adjusted stock selectivity
index" to compare educational attainment of emigrants with that of sending country's population
in 2011q1, assuming the same (country and arrival period-specific) age distribution as for the
stock of emigrants from this country to the given destination. This way, we are indeed measuring
"brain drain" rather than "diploma drain".
Data from the UK, the destination of about two-thirds (respectively, about three-fifths; one-
fifth) of Latvian (respectively, Lithuanian; Estonian) post-2000 expats in EEA/EFTA countries are
presented in Figure 8 (lower panel); other destinations are featured in Figure 9.
More than a half (respectively, almost a half; about 45%) of Estonian (respectively, Latvian;
Lithuanian) emigrants who arrived to the UK in the 21st century were tertiary-educated by 2011.
This is substantially more than among their age peers back home at the same time, as the
corresponding values of stock selectivity index (or the "brain drain index") are well above zero
(Figure 8, lower panel)xvi.
In other European OECD countries (excl. the German-speaking ones), university graduates
accounted, on average, for 30% of Latvian and 35% of Lithuanian emigrants (as of 2011), but just
for 20% of their Estonian colleagues. The "brain drain index" is negative (and falling) for Estonia,
while for Latvia and Lithuania it displays a positive trend being, on average, close to zero for pre-
accession and post-accession waves, but strongly positive during the crisis (Figure 9, top panel).
Overall, thus, by 2011, university graduates were over-represented among post-2000 Latvian
and Lithuanian emigrants in European OECD countries but under-represented among Estonian
emigrants. The extent of brain drain from the Baltics, however, varied strongly depending on
destination country (Figure 9, lower panel): the proportion of tertiary-educated among Estonian
emigrants is extremely low in Finland, moderate in Ireland and rather high elsewhere, while the
share of university graduates among Latvian and Lithuanian emigrants in Ireland is much lower
than in other European destinations.
20
On age-adjusted basis, the low-educated are over-represented among post-2000 Baltic
emigrants in OECD/Europe taken together, but the opposite is found for Estonian emigrants in the
UK, Estonian and Lithuanian emigrants in Ireland and Sweden and Latvian emigrants in Sweden,
see Figure 8 (lower panel) and Figure 9.
Figure 8. Skill composition and selectivity of emigrants from the Baltic countries, 2000-2013.
Top: Baltic mobile workers reported as household members in the home country (LFS data).
Bottom: Baltic-born residents of the UK aged 15+ (stock, 27.03.2011, by arrival time)
Notes: In the top panel, the (flow) selectivity index compares, for each period, "recent" mobile workers with
sending country's population aged 18-64 in the same period. In the bottom panel, the (stock) selectivity index is age-
adjusted, i.e. calculated vs. sending country's population in 2011q1, assuming the same age distribution as for the
stock of emigrants from this country to the UK (according to the UK Census data). Sources: Top - calculations with
LFS data. Bottom - calculation with the Population Census data (UK, and the Baltic countries) and Eurostat data.
0%
10%
20%
30%
40%
50%
-1.75
-1.25
-0.75
-0.25
0.25
0.75
2001-2003
2004-2008
2009-2010
2011-2012
2002-2003
2004-2008
2009-2010
2011-2012
2013
2002-2003
2004-2008
2009-2010
2011-2013
Labor migrants from
Estonia
Labor migr ants from Latvia
Labor migrants from
Lithuania
Selectivity index at departure
Share of tertiary-educated (right scale)
Low-educated selectivity index
Tertiary educated selectivity index
Share of low-educated (right scale)
0%
10%
20%
30%
40%
50%
60%
70%
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
2001-2005
2006-2009
2010-2011
2001-2005
2006-2009
2010-2011
2001-2005
2006-2009
2010-2011
Estonia-born UK residents,
27.03.2011, by arrival time
Latv ia-born UK residents,
27.03.2011, by arrival time
Lithuania-born UK residents,
27.03.2011, by arrival time
Stock selectivity inde x, age-adjusted
Share of tertiary-educated (right scale)
Low-educated selectivity index
Tertiary educated selectivity index
Share of low-educated (right scale)
21
Figure 9. Skill composition and selectivity of the 21st century emigrants from the Baltics.
Top: Total stock in OECD/Europe (excl. the UK and German-speaking countries), early 2011.
Bottom: Ireland, Sweden, Finland and Norway, early 2011.
Notes: The (stock) selectivity index is age-adjusted, i.e. calculated vs. sending country's population in 2011q1,
assuming the same (period-specific) age distribution as for the stock of emigrants (aged 15+) from this country to the
countries considered. Data were not available for Baltic emigrants in Germany, Austria and Switzerland. For each
destination country, its share in the stock of post-2000 emigrants from given Baltic country to OECD/Europe (as of
early 2011) is shown in parentheses. Sources: OECD (2014b), Eurostat data on population by educational attainment
level, sex and age, and own calculation.
For emigrants from each of the three Baltic countries in the UK, the highest value of the stock
SI for tertiary-educated, as well as the lowest value of the stock SI for the low-educated, is found
during the crisis period (Figure 8, lower panel). The same is true for Latvian and Lithuanian
0%
10%
20%
30%
40%
50%
60%
70%
80%
-0.75
-0.50
-0.25
0.00
0.25
0.50
0.75
1.00
1.25
2001-2005
2006-2009
2010
2001-2005
2006-2009
2010
2001-2005
2006-2009
2010
Estonia
Latv ia
Lithuania
Stock of Baltic emigrants in European OECD countries (excl. UK, Germany, Austria and Switzerland),
early 2011, by country of birth and arrival period
Stock selectivity inde x, age-adjusted
Share of tertiary-educated (right scale)
Low-educated selectivity index
Tertiary educated selectivity index
Share of low-educated (right scale)
0%
20%
40%
60%
80%
2001-2005
2006-2009
2010
2001-2005
2006-2009
2010-2011
2001-2005
2006-2009
2010
2001-2005
2006-2009
2010
2001-2005
2006-2009
2010-2011
2001-2005
2006-2009
2010
2001-2005
2006-2009
2010
2001-2005
2006-2009
2010-2011
2001-2005
2006-2009
2010
2001-2005
2006-2009
2010
Finland
(47%)
Ireland
(4%)
Norway
(7%)
Sweden
(9%)
Ireland
(12%)
Norway
(5%)
Sweden
(4%)
Ireland
(10%)
Norway
(10%)
Sweden
(4%)
Estonia
Latv ia
Lithuania
Stock of Baltic emigr ants on 01.01.2011 (Ireland: 26.04.2011), by country of birth, destination and arrival time
Share of tertiary-educated emigrants
Share of low-educated emigrants
Share of tertiary-educated in the sending country (age -adjusted)
Share of low-educated in in the sending country (age-adjusted)
22
emigrants in other European OECD countries, as well as for Estonian emigants in Ireland, Sweden
and (regarding high-educated) Norway (Figure 9, lower panel). This, once again, provides strong
empirical support to the idea that brain drain has intensified during the crisis (emigration from
Estonia to Finland being an exception).
The hypothesis that post-accession emigrants are less educated than the pre-accession ones
also seems to be consistent with the data from the main destination countriesxvii. This is the case
for Latvian and Lithuanian emigrants in the UK and Ireland, as well as for Estonian emigrants in
Finland and for emigrants from each of the Baltic countries in Sweden and Norway (see lower
panels of Figures 8-9).
We conclude the analysis of emigrants' selectivity on human capital using data from online
survey of the Latvian emigrants which attracted about 14 thousand respondents across the world
xviii. To save space, we focus on tertiary-educated. This time, we are able to compare selectivity at
departure and the stock selectivity (using age-adjusted indexes in both cases). The data, covering
emigrants which left Latvia in 2000-2014 are presented in Figure 10, separately for Ireland, the
UK and other EU/EFTA countries, as well as for all EU/EFTA destinations.
Figure 10 Shares, flow selectivity and stock selectivity of tertiary educated emigrants from
Latvia, 2000-2014, by destination and arrival period
Notes: The (stock) selectivity index compares, for each destination and arrival period, human capital of the
stock of adult Latvian emigrants in August - October 2014 with human capital of Latvia's population as of 2014,
assuming the same (destination-and-arrival-period-specific) age distribution as for the stock of emigrants. The
(flow) selectivity index compares, for each destination and arrival period, the share of emigrants who left Latvia
aged 15+ with completed tertiary education with the share of tertiary educated stayers in that period, assuming
the same age distribution as for those who moved from Latvia during that period. Sources: Calculation with
emigrant survey data and Eurostat data.
0%
10%
20%
30%
40%
50%
60%
-0.3
-0.2
-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
2000-2003
2004-2008
2009-2011
2012-2014
2000-2003
2004-2008
2009-2011
2012-2014
2000-2003
2004-2008
2009-2011
2012-2014
2000-2003
2004-2008
2009-2011
2012-2014
2000-2014
Ire land
UK
Oth er EU/ EFTA
EU/EFTA total
Emigrants fr om Latvia (stock as o f Aug-Oct 2014), by arrival period
Selec tivity index, age-adjusted
Share of tertiary-
educated among
adult emigrants by
2014 (right scale)
Tertiary educated:
stoc k se lec tivity
index, 2014
Tertiary educated:
selec tivity index at
departure
Share of tertiary-
educated in L atvia
among adult
emigrants at
departure (right scale)
23
Noteworthy, choice of destination was not neutral with respect to human capital: Between
2000 and 2014, the proportion of university graduates among adult Latvian emigrants arriving to
Ireland (respectively, the UK and the rest of EU/EFTA) varied from 13% to 26% (respectively,
from 34% to 37% and from 21% to 37%). Except for Ireland in 2000-2008, these shares were well
above those found in respective periods among stayers of the same age as emigrants, as indicated
by positive values of selectivity index at departure (ranging between 0.23 and 0.48 for the total
outflow to EU/EFTA). Thus, these data confirm substantial (and increasing over time) diploma
drain from Latvia to various EU/EFTA destinations during the whole period between 2000 and
2014. The share of tertiary educated among emigrants further increased during their stay in the
host countries, reaching, by mid-2014, 45% on average across destinations and arrival periods
(Figure 10; note that the shares shown for 2014 in the Figure refer to those aged 15+ in 2014, but
they would change by less than 1 percentage point if those who were aged 15+ when arrived
would be considered). Ireland aside, the brain drain selectivity index (stock SI) takes positive
valuesxix, indicating that by 2014, the share of university graduates among Latvian emigrants in
each of the destinations under inspection was higher than among their age peers in Latvia.
Do data presented in Fugure 10 support our expectations regarding selectivity patterns? Our
first hypothesis stated that the post-accession emigrants are, as a group, less educated than the pre-
accession ones. This is strongly supported by selectivity index at departure for all destinations
considered, as well as by stock selectivity index in Ireland and the UK. Moreover, for the UK this
is also true in terms of the shares of university graduates among emigrants found in 2014.
Regarding the crisis period we had ambiguous expectations: Baltic residents with higher
education were, on average, less hardly than others hit by unemployment but, plausibly more
likely to emigrate for non-economic reasons. Evidence from Figure 10 (the share of university
graduates at departure) suggests that during the crisis, diploma drain from Latvia was more
intensive than before, although for destinations other than Ireland this reflected fast growth of
high-educated segment of the young and middle-aged population in Latvia, while the selectivity
index was falling. On the other hand, the stock slectivity index suggests that by 2014, the crisis
wave (and, except for the UK, also the post-crisis wave) of emigrants from Latvia featured more
intensive brain drain than the post-accession wave.
Hazans (2012: 193-194) provides econometric analysis of determinants of work abroad using
data from Estonian and Latvian LFS 2001-2009. He finds that propensity to work abroad among
the university graduates (respectively, among the low-skilled) has decreased (respectively,
increased) after EU accession both in Estonia and Latvia. During the first year of the crisis, the
differences between the skill groups in terms of propensity to work abroad have disappeared in
Latvia (supporting the idea of systemic nature of the crisis there), while in Estonia these
differences hardly changed since the pre-crisis period.
Yet another way of measuring the brain drain is suggested in Hazans (2013c: Figure 4.21),
where it is shown that during the crisis emigrants from Latvia (especially the top half) were
relatively much more productive than before. This supports the hypothesis that the brain drain
intensified during the crisis. This effect, however, did not last for more than two years.
24
Ethicity and citizenship
Figure 11 (panels A-C) presents empirical evidence from Latvia and Estonia on ethnic
composition of the four recent waves of emigrants. As seen in panel A (which extends results of
Hazans 2012: Section 5), the share of minorities among the individuals working abroad but still
considered household members at home is U-shaped, reaching its minimum in 2006-2008 (when
the ethnic gaps in employment and unemployment in both countries were at their lowest values,
see Figure 7) and increasing during the crisis, when relative labor market position of ethnic
minorities deteriorated. Corresponding selectivity index (which accounts for the fact that minority
share in the population of the sending countries was declining over time and is smaller among the
youth and the middle-agers than among the elderly) follows the same pattern, consistent with
expectations stated in Section 5.xx Moreover, as expected, selectivity indexes of minority
individuals with and without Estonian/Latvian citizenship move in opposite directions (except for
the second part of the post-accession period).
The Latvian results are fully consistent with the expectations: after accession, propensity to
work abroad increased among minority population with Latvian citizenship but decreased among
non-citizens and those holding CIS countries' passports. During the crisis it was the other way
around (both in Latvia and Estonia); in other words, the crisis-triggered joblessness was strong
enough push factor to overcome the institutional barrier - lack of free mobility provisions for
residents of Latvia and Estonia without citizenship of these countries. In the post-crisis period,
however, propensity to work abroad is again higher among minority citizens than among
noncitizens.
25
Figure 11 Ethnic composition and selectivity of emigrants from Estonia and Latvia,
2000-2013
Notes: In the Latvian part of panel A, citizenship-specific selectivity indexes for 2000-2003 due to data
limitations refer to 2002-2003. In panel B, the data refer to persons which where either officially employed or
registered unemployed in Latvia for some time in 2005-2011, have neither died nor retired in Latvia but have
disappeared (at age below 60) at least for a year from both State Social Insurace Agency (SSIA) and State
Employment Agency (SEA) records. In panel C, the data com from the UK Population Census 2011 but cover
only England and Wales. Sources: Panel A: National LFS data and own calculation. Panel B: Merged records of
SSIA, SEA and Population Register and own calculation. Panel C: ONS (2014a) and own calculation.
Noteworthy, Latvian minorities were over-represented among mobile workers still attached to
their Latvian households in the whole period between 2000 and 2013 (selectivity index falls from
0.42 before accession to 0.06 in 2006-2008 and rises again to 0.25 in 2011-2013), while their
Estonian counterparts were under-represented in 2004-2012, as seen by negative values of
selectivity index. Plausibly, the Estonian case can be explained by the comparative advantage
ethnic Estonians have over Russians in Finland's labor market due to similarity between Estonian
and Finnish languages. However, the share of Russian-speakers among Estonia-born in Finland
seems to be on rise since 2008 (Hazans 2014b:12).
Panels B of Figure 11 is based on Latvian administrative data and refers to emigrants which
left Latvia in 2005-2011 and have been officially employed and/or registered as unemployed in
Latvia during this period. Panel C of Figure 11 is based on the data of the UK Population Census
2011 and refers to Latvia-born residents of England and Wales which arrived to the UK in 2000-
2011 (before the Census). For 2005-2011, these two independent data sources (both free from the
restriction that the emigrants are still considered household members in Latvia) give consistent
estimates of the proportion of non-Latvians among emigrants: between 49% and 40% in general
and between 44% and 43% in England and Wales; both sources suggest that this proportion was
falling over time, yet minorities remained over-represented among emigrants. Panel C supports
our expectation (see Section 5) that the proportion of ethnic minorities among post-accession
emigrants (and respective selectivity index) was smaller than before. Unlike the Latvian part of
0%
10%
20%
30%
40%
50%
60%
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
2001-2003
2004-2005
2006-2008
2009-2010
2011-2012
2000-2003
2004-2005
2006-2008
2009-2010
2011-2013
2005
2006-2008
2009-2010
2011
2000-2003
2004-2008
2009-2011
Estonia
Latvia
Latvia
Latvia
A: L abor migrants reported as house hold members in
the sending country (LFS data, by observation year)
B: Emigran ts with
labor marke t
experience in Latvia,
by year of departure
C: Emigran ts in
the UK, stoc k on
27.03.2011, by
arrival time
Selectivity index at departure , age-adjusted
Share of minority (right scale)
Minority selectivity index
Minority-citizens selectivity index
Minority-other selectivity index
26
panel A, panels B and C do not feature an increase in the minority selectivity index after the crisis;
this might have to do with the data coverage (only (part of) 2011 in panel C as opposed to 2011-
2013 in panel A) but also with different definitions of "emigrants".
Note that even when ethnicity and citizenship are controlled for, country of birth is an
important driver of emigration: other things equal, foreign-born minority individuals are much
more likely to work abroad (Hazans, 2012: 193-194).
Gender and familyxxi
We start this subsection by looking at gender balance of emigration flows across time periods
and European destinations (Figure 12). The share of females in total gross outflow from Estonia
(respectively, Latvia, Lithuania) to EU/EFTA countries was declining over time: from 62%
(respectively, 57%, 54%) in the pre-accession period to 54% (respectively, 46%, 47%) in 2009-
2013. What are likely explanations for higher female share in pre-accession emigration? First, in
the absence of free movement of labor, non-labor migration accounted for a larger share of
outflows: according to the survey of Latvian emigrants (see endnote xviii), 25% of females which
moved from Latvia in 2000-2003 indicate finding a foreign spouse/partner as one of the reasons
for emigration, while this proportion falls to 14% in 2004-2008 and 11% in 2009-2013. Second,
while at that time there was no easy ways to find jobs in formal sector abroad, working au pair
(often informally) was a popular first step available almost exclusively to females.
Figure 12 Female share in gross outflows from the Baltic countries by destination, 2000-2013
Source: Calculation with data of Eurostat and UK Department of Work and Pensions.
Another noteworthy message from Figure 12 is that the main destinations of labor migrants
(Finland for Estonia, the UK and Ireland for Latvia and Lithuania) feature roughly equal
0%
10%
20%
30%
40%
50%
60%
70%
80%
UK
Ire land
Germany
Finland
Norway
Other Nordic
Oth er EU/EFTA
Total EU/EFTA
UK
Ire land
Germany
Norway
Other Nordic
Oth er EU/EFTA
Total EU/EFTA
UK
Ire land
Germany
Norway
Other Nordic
Oth er EU/EFTA
Total EU/EFTA
Estonia
Latvia
Lithuania
2000-2003
2004-2008
2009-2013
27
proportions of males and females. By contrast, flows to some of the other important destinations
remained female-dominated also in the post-accession period; this was the case for the flows to
Germany before it opened its labor market in 2011 (and also after that for Estonia), as well as for
Estonian, Latvian and (during and after the crisis) Lithuanian flows to destinations other than the
UK, Ireland and the Nordic countries. On the other hand, post-accession flows to Norway from all
Baltic countries were male-dominated (likely due to high demand for physically demanding
occupations).
Table 2 illustrates that shift towards family migration during and after the Great Recession
(see Table 1) increased the share of children in Latvian and Lithuanian (respectively, Estonian)
outflows to the Nordic countries from about 10% (respectively, 14%) in 2004-2008 to 14%
(respectively, 18%) in 2009-2013, while in outflows from Latvia and Lithuiania to Ireland this
share increased from 11% in 2007-2008 to about 20%in 2009-2013. In the same time, number of
Baltic children left in foster care by emigranted parents also increased substantially: in Latvia, the
such children accounted for 5% to 6% of all students of general schools in 2010-2013xxii ; in
Lithuania, the annual number of new requests for foster care for to parents’ emigration in 2008 -
2011 has more than doubled compared to the pre-crisis level (Sipavičiene and Stankuniene, 2013:
Table 3.4). This problem attracted a lot of public attention; special government regulations have
been adopted in Latvia and Lithuania (Broka, 2011; Sipavičiene and Stankuniene, 2013).
Table 2 Share of children below 15 years of age in emigration flows from the Baltic countries
to selected destinations, by period
Outflows to the Nordic countries
Outflows to Ireland
Estonia
Latvia
Lithuania
Latvia
Lithuania
2000-2003
13.9%
10.3%
6.0%
2004-2008
13.6%
9.9%
9.5%
2007-2008
10.9%
11.4%
2009-2013
18.3%
14.0%
13.6%
2009-2013
21.0%
18.2%
Source: Calculation with Eurostat data
According to to the survey of Latvian emigrants (see Hazans (2015a) and endnote xviii), 10%
of adult Latvian emigrants have children of pre-school or school age left in Latvia, while 25%
live with such children abroad.
7 Emigrants' labor market status, occupation and downskilling
Figure 13 (top panel) compares employment , unemployment and inactivity rates of Baltic
emigrants and natives in the UK, Ireland (for Estonia - Finland) and other European OECD
countriesxxiii in 2011, separately for the youth and the 25-64 years olds (we refer to the latter group
as "adults" hereafter); the corresponding rates for stayers in the Baltic countries are shown
alongside.
In the UK labor market, Baltic emigrants fare much better than natives (and of course than
stayers in their home countries). Among the youth, emigrants enjoy much higher employment and
substantially lower unemployment rates. Baltic-born adults in the UK feature unemployment rates
similar to those of natives (but lower than those found in the Baltics). Employment rates of
Latvia- and Lithuania-born exceed those of natives (respectively, stayers) by10 (respectively, 15)
28
points, while inactivity rates are by 8 to 10 points lower than among natives and stayers. For
Estonians, the gaps are somewhat smaller but still substantial.
In Ireland, employment rates of Baltic-born youth (about 50%) are lower than in the UK, yet
well above those of Irish-born youth (30%), let aside Baltic stayers; youth unemployment rates
are high (around 25%) yet lower than 30% found among both Irish-born youth in Ireland and
Baltic youth at home. Situation among adults differs: employment rates of Latvians and
Lithuanians in Ireland (almost 70%) are similar to those of native-born and also to those found in
the home countries, but unemployment rates among the Baltic-born, at 24%, are twice as high as
among natives and by about 10 points higher than among stayers. Inactivity rates among the
Baltic emigrants in Ireland, at about 10%, are much lower than among natives (25%) and Baltic
stayers (17%).
The main destination of Estonian emigrants is Finland; here, Estonia-born youth feature
similar employment rate but lower unemployment rate than their native-born counterparts; on
both accounts young emigrants outperform young stayers. By contrast, among Estonia-born adults
in Finland employment rate is somewhat lower, but unemployment - twice as high as among
natives; moreover, among adult stayers both indicators are better than among emigrants in
Finland. Like in case of Latvians and Lithuanians in Ireland, this disadvantage is outweighed by
higher earnings and benefits in Finland.
Finally, in other European OECD countries young Baltic emigrants have, on average, higher
employment rates than their counterparts in home countries. On the other hand,employment rates
of Baltic emigrants aged 25-64 are, on average, 5 to 10 points below those of stayers in the same
age.
Figure 13 Labor market status of Baltic emigrants and natives in selected European OECD
countries in comparison with Baltic stayers, 2011
Top: Youth and adult employment, unemployment and inactivity rates
Bottom: Employment gaps by gender and educational attainment (age 15-64)
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
Latv ia
Lithuania
Estonia
Natives
Latv ia
Lithuania
Natives
Estonia
Natives
Latv ia
Lithuania
Estonia
Latv ia
Lithuania
Estonia
Latv ia
Lithuania
Estonia
Natives
Latv ia
Lithuania
Natives
Estonia
Natives
Latv ia
Lithuania
Estonia
Latv ia
Lithuania
Estonia
UK
Ire land
Finland
Other (ex.
DE, AT, CH)
UK
Ire land
Finland
Other (ex.
DE, AT, CH)
Baltic-born vs native-born, by host country
Staye rs
Baltic-born vs native-born, by host country
Staye rs
age 1 5-24
age 2 5-64
Emplo yment
Unemployment
Inac tivity
29
Notes: "High" refers to teriary education (ISCED11 levels 5-8); "Medium" - to upper secondary and post-
secondary non-tertiary education (ISCED11 levels 3-4); "Low" - to ISCED11 levels 0-2 (lower secondary, primary
and below). Sources: Emigrants - calculations with data from OECD (2014b). Natives and Baltic stayers - calculation
with EU LFS data (Eurostat).
As shown in the bottom panel of Figure 13, tertiary-educated Baltic emigrants enjoy
employment rates by 6 to 9 points higher than their medium-educated counterparts in all main
European destinations; these skill-driven employment gaps among emigrants are, however,
substantially smaller than among natives in the same countries and among Baltic stayers.
Employment gaps between medium- and low-educated Baltic emigrants are absent in the UK and
Finland (where such gaps for natives are very large); by contrast, in Ireland and other European
destinations employment rates of medium-educated emigrants exceed those of low-educated
emigrants by 14 to 20 points, yet these gaps are slightly (respectively, much) smaller than those
found among natives (respectively, Baltic stayers). We conclude that employment gain from
emigration is inversely related to education level among Baltic emigrants.
While gender gaps in employment are close to zero among stayers in Latvia and Lithuania,
among emigrants from these countries in their main destination, the UK, male employment rates
exceed female ones by 7.5 and 4.5 points in Latvian and Lithuanian case, respectively; these gaps
are smaller than 10 points found among UK natives. In other destinations, gender gaps in
employment are close to zero among Lithuanian emigrants and vary between 2 and 5 points
among their Latvian counterparts, well below the gender gaps among natives. Estonian male and
female emigrants feature almost equal employment rates in all destination countries, while among
stayers in Estonia male employment rate is by 5 points higher (Figure 13, bottom).
More detailed information available from UK Population Census 2011 (ONS 2013, 2014a)
indicates that 20% of employed Lithuanian emigrants in England and Wales were self-employed;
this proportion is similar to that found among emigrants from non-Baltic NMS but higher than
among natives (15%) and Latvia-born emigrants (9%); note that in the same year there were about
11% self-employed among employed stayers in both Lithuania and Latvia.
-10
0
10
20
30
40
50
-5
0
5
10
15
20
25
Latv ia
Lithuania
Estonia
Natives
Latv ia
Lithuania
Natives
Estonia
Natives
Latv ia
Lithuania
Estonia
Natives
Latv ia
Lithuania
Estonia
UK
Ire land
Finland
Oth er ( ex. D E, AT, CH)
Baltic-born emigrants vs native-born, by host country
Baltic stayers
Employment gap, % points
Employment gap, % points
High vs Medium education
Men vs Women
Medium vs Low education (right scale)
30
Figure 14, based on a survey conducted in 2014xxiv , provides a more detailed breakdown of
post-2000 Latvian emigrants' main activities abroad (by gender, education, destination country
and period of arrival). Overall, two-thirds are employees, about 7% - self-employed, 9% -
students, 3% - jobseekers, 10% care for children or other family members, and just 5% are other
inactive (note that 99.5% of respondents are aged 15-64). Employment rate is much higher among
men (83% vs 64% among women), but this is largely balanced by child (and other family
members) care duties (16% among women vs. 3% among men). While recent emigrants feature
somewhat lower employment rates (72% and 68% among those who arrived in 2009-2011 and
2012-2014, respectively, vs. 80% among those who arrived in 2004-2008), the proportion of
emigrants who are either employed or students is more stable: 82%, 85% and 81% for pre-
enlargement, post-enlargement and post-crisis waves, respectively. Proportion of self-employed
increases with education level; it is lowest in Ireland and highest in non-English speaking
EU/EFTA destinations, as well as in destinations outside EU/EFTA.
Regarding emigrants' occupation, we use two complementary data sources (see Figure 15).
Data of UK Population Census 2011 (ONS 2014b) cover all emigrants considered as usual
residents as of March 27, 2011; for comparison, data on UK natives and Baltic stayers are
presented alongside. In addition, for Latvian emigrants we provide data by period of arrival to the
UK. On the other hand, Latvian and Lithuanian LFS data (2000-2013; split into three periods)
provide information on residents of these countries working abroad but still considered household
members at home.
According to both data sources, Baltic emigrants and mobile workers, in comparison with
stayers (and in the UK, also with natives) feature much smaller proportions of high-skilled
nonmanual workers and much larger proportions of elementary occupations (Figure 15). When
split by gender (these results are not shown in the Figure), it appears that female emigrants and
mobile workers, in comparison with their male compatriots, feature by 3 to 5 points higher
proportions of high-skilled nonmanual occupations, much higher proportions of low-skilled
nonmanual (clerical, sales and service) occupations and much lower proportions of skilled manual
occupations; the share of elementary occupations among Estonian and Latvian emigrants (as well
as among post-crisis Latvian mobile workers) does not vary by gender, but is much higher among
Luthuanian female emigrants and mobile workers than among their male counterparts.
Next we discuss how does emigrants' occupational profile change over time and with
duration of stay in the host country. First, we might observe wave-composition effects related to
fall in emigrants' human capital after EU enlargement of 2004, and country-specific changes in
emigrants' human capital after the start of the Great Recession at the end of 2008 and with the
beginning of recovery in 2011 (see sections 5 and 6). Second, one might expect wave-specific
behavioral effects, especially among post-2008 emigrants (see section 5 for discussion of the
context): those who left home countries during the crisis were likely to be less choosy with respect
to their job abroad; on contrary, the post-crisis emigrants (and even part of the crisis-time ones),
being interested in long-term or permanent emigration could postpone moving until finding a job
where they can to some exten use their skills and/or hope for promoting in future. Third, skill-
composition of host countries' labor demand might also change over time (recently, such a change
was "pro-skill" in most cases). Fourth, steady growth of the Baltic diasporas (especially since
2004, see Figure 3), suggests that migrant network effect resulted in better employment
opportunities for more recent emigrants. Finally, longer time spent by emigrants in the host
31
country is often associated with better local language skills, increased social capital, larger general
and/or specific experience and, hence, with better chances to be promoted or to find a job which
better matches their qualifications (if downskilling was accepted in the first place).
In the case of UK Census data, different waves of emigrants are observed at the same time, so
both time and duration of stay effects are at work. According to the data presented in Figure 15,
post-acession Latvian emigrants feature "less skilled" occupational profile than pre-accession
ones, and this profile further deteriorates after the onset of the crisis, suggesting that negative
effects (duration of stay, as well as some wave-composition and behavioral ones) were stronger
than the network effects and and pro-skill changes in the UK labor demand.
Internationally-mobile workers found in national LFS data are assumed to be absent from
home country for less than a year xxv, hence occupational profiles based on these data are subject
to time effects but not to duration effect. Indeed, both Latvian and Lithuanian post-accession
mobile workers abroad feature lower proportions of all nonmanual and high-skilled nonmanual
occupations than pre-accession ones (Figure 15). On the other hand, Latvian (respectively,
Lithuanian) mobile workers observed during and after the crisis (in 2009-2013) have the same
(respectively, higher) occupational profile than their pre-crisis counterparts, which is consistent
with increase in shares of both high- and low-skilled among emigrants (see section 6) and
presence of behavioral, network and demand effects working in opposite directions.
Figure 14. Post-2000 emigrants from Latvia: main activity abroad in 2014,
by educational attainment, destination and period of arrival
Source: Calculation with emigrant survey data (see endnote xviii) , N=11058 (after excluding non-response of 11.9%).
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Men
Women
Low
Medium
High
UK
Ire land
Othe r EU/ EFTA
Othe r
2000-2003
2004-2008
2009-2011
2012-2014
Total
Gender
Educ ation
Des tination
Period of arrival
Other inactive
Child care or family member care
Job se eking
Studi es/Tra ining
Self-employme nt
Wage employment
32
Figure 15. Employed emigrants from the Baltic countries and stayers therein by occupation
Source: Calculation with the data of the UK Population Census 2011 (ONS 2014 a, b, c) and LFS data.
An important question is how many of emigrants experience downskilling, i.e. are
overqualified for the jobs they hold abroad. We look at this question from three perspectives.
First, we use the standard OECD definition of overeducation: an employed person with tertiary
(respectively, upper secondary) education is overeducated if he/she works in manual or low-
skilled nonmanual (respectively, elementary) occupation. Second, we compare emigrant's
occupation with his/her last occupation in the home country. Following the spirit of the OECD
definition, those who moved from high-skilled nonmanual to other occupations have experienced
downskilling, as well as those who moved from low-skilled nonmanual or skilled manual to
elementary occupations (reversed moves will be referred to as upskilling). Third, we use answers
to the question "Do you to a large extent use your education/qualification in your job?" from
survey of Latvian emigrants (negative answers are interpreted as self-assessed downskilling). The
results are summarised in Figure 16 for the stock of Latvian emigrants as of 2014, as well as for
Latvian and Lithuanianxxvi mobile workers abroad and stayers observed in 2000-2013.
In the period under inspection overeducation rates among employed stayers varied in a
narrow range: 13% to 16% in Latvia 11% to 13% in Lithuania, with a slight increase during and
after the crisis; among university graduates about 20% were overeducation in both countries
(Figure 16, top). By contrast, mobile workers abroad feature higher overeducation rates, and in
2009-2013 this rates declined compared to the pre-crisis period: from 25% to 21% (respectively,
from 35% to 25%) among all mobile workers from Latvia (respectively, Lithuania) and from 41%
to 39% (respectively, from 66% to 54%) among those with higher education. Furthermore,
downskilling vs. the last job at home among Latvian mobile workers also declined from 12% in
2004-2008 to 6% in 2009-2013. Migrant network effect, increased preference for long-term
emigration and changes in demand structure (discussed earlier in this section) are likely reasons
why, despite the crisis, recent Latvian and Lithuanian mobile workers were able to find jobs
matching their qualifications more often.
About one-third of employed Latvian emigrants in 2014 were formally overeducated, and the
same proportion experienced downskilling vs. their last occupation in Latvia, while almost 60%
0%
20%
40%
60%
80%
100%
before 2004
2004-2008
2009-2011
total
Lithuania, total
Estonia, total
Native s
Latv ia
Lithuania
Estonia
2000-2003
2004-2009
2009-2013
2002-2003
2004-2009
2009-2013
Latv ia
.
.
.
Employed population in UK, 2011, by
country of bir th and ar rival time
Employed Baltic
stayers, 2011
Latvian mobile
workers abroad
Lithuanian mobile
workers abroad
Elementar y
Skilled manual
Low-skilled nonmanual
High-skilled nonmanual
33
reported little or no use of their education or qualification (Figure 16, bottom). Among all
emigrants and mobile workers alike, all measures of brain waste are substantially higher among
women (Figure 16, top and bottom).
In the context of brain waste, the largest concerns are usually associated with college
graduates. Noteworthy, while this category of emigrants (and mobile workers) feature higher
overeducation rates than their medium-educated counterparts, it is the other way around for both
downskilling (vs. last job at home) and self-reported skill underutilization (Figure 16, top and
bottom). Furthermore, while about 40% of employed high-educated Latvian emigrants are
overeducated and the same proportion report skill underutilization, only 24% are downskilled vs.
their last occupation in Latvia (Figure 16, bottom); this difference is even more pronounced (35%
vs. 1%) among high-educated mobile workers (Figure 16, top). In other words, a large part of
overeducated tertiary-educated emigrants and mobile workers were overeducated already in
Latvia.
The extent to which emigrants are subject to downskilling strongly varies across destination
countries: All measures of brain waste among Latvian emigrants take their highest values in the
UK and Ireland, followed by Germany and the Nordic countries, followed by other EU/EFTA
countries, while the lowest values are found in destinations outside EU/EFTA (Figure 16, bottom).
34
Figure 16. Various measures of brain waste among Baltic emigrants and stayers.
Top: Latvian and Lithuanian mobile workers and stayers, 2000-2013, by observation period
Bottom: Employed Latvian emigrants as of 2014, by arrival period
Notes: For Latvia, data on downskilling refer to 2002-2013. For Lithuania, results on downskilling are absent due to
data limitations. Sources: Top - calculation with Latvian and Lithuanian LFS data. Bottom - calculation with emigrant
survey data (see endnote xviii) .
0%
10%
20%
30%
40%
50%
60%
70%
2000-2003
2004-2008
2009-2013
Medium
High
Men
Women
2002-2003
2004-2008
2009-2013
Medium
High
Men
Women
Period
Education
Gender
Total
Period
Education
Gender
Total
Latvia
Lithuania
Mobile workers: Overeducated
Mobile workers: Downskilled vs. occupation in Latvia
Mobile workers: Overeducated among univer sity graduates
Stayers: Overeducated
Stayers: Overe ducated among university graduates
0%
10%
20%
30%
40%
50%
60%
70%
80%
UK & Ire land
Germany & Nordic
Oth er EU/EFTA
Other
2000-2003
2004-2008
2009-2011
2012-2014
Low
Medium
High
Men
Women
Destination
Arr ival period
Education
Gender
Total
Employed Latvian emigrants, as of 2014
downskilled vs. occ. in Latvia
upskilled vs. occ. in Latvia
overeducated
overed. university graduates
no or little use of one's
education/qualification
35
8. Return migration, migration networks and intentions to return
Figure 17 compares flows of return migrants to the Baltic countries before, during and after
the crisis. The returning flows are measured as percentage of the outflows of nationals or
respective country in the same year.
In 2005-2008, the patterns or return were similar in all three countries (reflecting similar
positive developments in their economies in up until the first half of 2008): returning flows
initially accounted to about 30% of the outflows; until 2007 this "return rate" remains almost
stable (except for Lithuania where it exceeds 40%), but in 2008 exceeds 50% in Latvia and 60%
in Lithuania; a smaller increase (to 36%) is found in aslo in Estonia (recall that the main
destination country, of Estonian emigrants, Finland, is geographically and linguistically very
close, which makes pressure to return smaller). The sharp increase in return in 2008 was likely
caused by combination of previous increase in earnings and fall of unemployment in the home
countries (see Figure 1) with the first signs of the crisis in receiving countries and depreciation of
the British pound and the Nordic currencies.
During 2009-2010, the Lithuanian and Latvian return rates fell, respectively, by two-thirds and
by more than a half, reaching 18% for Lithuania and 23% for Latvia. In 2011-2012, as the Baltic
economies resumed growth, the number of Lithuanian and Latvian returnees increased faster than
the number of emigrants, reaching, respectively, 45% and 38% of the outflows. Estonia, as
already mentioned, is a special case; its return rate behaved similarly to the Latvian one but was
less volatile.
When the flows of return migrants to the Baltic countries from the main European destinations
are compared, it appears that during the crisis, the intensity of return flows from the UK, Germany
and the Nordic countries to Latvia and Lihuania (as well as to Estonia from Germany and the
Nordic countries) declined in relation to opposite emigration flows, while relative intensity of
return flows from the French-, Dutch- and German-speaking countries (excl. Germany) was less
volatile; it decreased for Latvia and Lithuania but increased slightly for Estonia. Return flows
from Russia to the Baltic countries became smaller relative to outflows.
As far as return migration from countries hardly hit by the crisis (the Southern member states
and Ireland) is concerned, relative intensity of return flows increased explosively and stayed very
high for at least four years; in 2011- 2012, the number or returnees from these destinations to
Latvia and Lithuania exceeds the number of emigrants.
Figure 17 presents also the share of returnees (defined as those who spent at least 6 months
working abroad over the last 10 years, lived in the country before that and returned) in population
aged 18-74 for each of the Baltic countries. This share was pretty stable both in Latvia (about 6%
in 2007-2010) and Lithuania (about 7% in 2009-2012). In Estonia, returnees accounted for 7% of
population already in 2004 (Finland nearby, with a similar language, played its role here); this
share varied around 9% in 2007-2011 and fell to 8% in 2012. Given the declining populations (see
Figure 3), it appears that in 2007 - 2012 the absolute stock of returnees was declining in Estonia
and Latvia and roughly stable in Lithuania, despite steady positive inflowsxxvii. This means that
substantial part of returnees repeatedly moved abroad. Indeed, according to a survey conducted in
Latvia in late 2010, 43% of the return migrants plan to move abroad again in the near future and
36
Figure 17 Inflow and stock of return migrants in the Baltic countries, 2004-2012
Notes: Inflow of return migrants from EU and OECD countries is expressed as % of the outflow of the sending
country's nationals in the same year. Stock of return migrants in population aged 18-74 is measured as % of persons
who spent at least 6 months working abroad over the last 10 years, lived in the country before that and returned.
Lithuanian stock datashown in 2012 refer to 2013. Sources: Inflow - Hazans (2015b), based on receiving countries
data (sending countries' data result in similar dynamics). Stock - calculation with data of European Social Survey
(rounds 2-6) and (Latvia, 2010) survey "National Identitty: Place, Capability, Migration" (see Hazans 2011b).
Figure 18. Stock of returnees in the Baltic countries, 2004 -2013:
Skill composition and selectivity with respect to stayers
Notes: See Notes to Figure 17 for the definition of the stock of return migrants. Selectivity index is age-
adjusted, i.e. calculated vs. country's population in the same year, assuming the same age distribution as for the
stock of return migrants. Sources: calculation with data of European Social Survey (rounds 2-6) and (Latvia,
2010) survey "National Identitty: Place, Capability, Migration" (see Hazans 2011b).
4%
5%
6%
7%
8%
9%
10%
10%
20%
30%
40%
50%
60%
70%
2004
2005
2006
2007
2008
2009
2010
2011
2012
stock vs. population aged 18-74
Inflow vs. outflow of nationals
inflow Estonia
inflow Latvia
inflow Lithuania
stoc k Eston ia
stoc k La tvia
stock Lithuania
0%
10%
20%
30%
40%
-0.75
-0.50
-0.25
0.00
0.25
2004
2007
2009
2011
2012
2007
2009
2010
2009
2011
2013
Estonia
Latv ia
Lithuania
Stock of return migrants in the Baltic countries
Stoc k selec tivity index, age -adjusted
Share of tertiary-educated (right scale)
Low-educated selectivity index
Tertiary educated selectivity index
Share of low-educated (right scale)
37
another 23% admit such a possibility (Hazans 2011b, p.98 and Box 2.33); furthermore, 48% of
post-2000 emigrants from Latvia moved from their home country more than once, according to
the survey of Latvian emigrants conducted in 2014 (see Hazans 2015a , 2015c).
Figure 18 compares education of return migrants with that of stayers using age-adjusted stock
selectivity index (see section 6 for details). In all three countries, the proportion of tertiary
educated among the returnees is either lower or (in Estonia, 2004-2007 and Latvia, 2009) just
slightly higher than among similarly aged stayers, suggesting that in terms of formal education
brain gain due to return migration is, at best, modest; however, the proportion of low-educated is
in most cases also lower than among stayers. See Hazans (2012: Table 6.6) for similar analysis of
the flows of returnees to Latvia and Estonia.
Apart from obvious positive demographic impact of return migration and (admittedly, modest)
gain it might bring in terms of educational attainment, there are other potential benefits the Baltic
countries can enjoy if more emigrants come back. The returnees bring useful foreign work
experience which manifests itself in higher earnings, other things equal (Hazans, 2008; Hazans,
2013c: 89)xxviii. According to the results of the survey concucted for the Latvian Human
Development Report 2011, 70% of returnees report that foreign experience had a positive effect
on their professional skills and 82% notice a positive effect on self-confidence (Hazans, 2013c:
88). Pungas et al. (2012) for Estonia provide evidence that education obtained abroad improves
the socialisation prospects later on.
To conclude this section, we briefly summarise findings from (Hazans, 2015a) on post-crisis
return intentions of Latvian emigrants, based on a survey conducted in 2014. Within 5 years 16%
of emigrants will (definitely or probably) return (this includes less than 4% planning to return
within 6 months). Another 16% plan or consider returning upon retirement. These findings are in
striking contrast with the situation observed in 2005-2006, when two-thirds of emigrants having
left Latvia in 2004-2005 were planning to return within two years, most of them (almost half of all
emigrants) even within one year (Hazans and Philips, 2010, Figure 9). Similarly to Pungas et al.
(2012), Anniste and Tammaru (2014) for Estonia, we find that other things equal, Russian-
speakers are less likely to return than ethnic Latvians, while those not using their
education/qualification at work are more likely to return. The effect of higher education on
propensity to return is negative for men but positive for women.
9. The economic impact of emigration
Emigration may affect the sending country’s labor market in a number of ways. First, it tends
to reduce unemployment below the levels expected under a zero-emigration scenario, because
actual or potential unemployed, and economically inactive individuals move abroad or fill the
vacancies left behind by previously employed emigrants. Table 3 (based on LFS data) indicates
that in 2003-2013 one-fifth to one-third of Latvian guestworkers experienced unemployment or
economic inactivity in Latvia during the year prior to their departure; the share of those coming
from unemployment was especially high during the crisis.
38
Table 3 Unemployment or spells of economic inactivity in Latvia during the year prior to departure,
among Latvian guestworkers (2002-2013)
Percentages
2002-2003
2004-2008
2009-2011
2012-2013
Unemployed
12.2
21.7
27.8
17.5
Inactive
17.3
13.8
8.0
5.6
Total
27.5
32.4
32.4
20.8
Notes: Guestworkers here are labor emigrants still considered household members back home.
Source: Calculations based on Latvian LFS data.
Moreover, among all emigrants with legal work or registered unemployment experience in
Latvia, the share of those whose last registered activity before leaving was unemployment, rose
from 10% in 2005 to 48% in 2011 (Figure 19).
Figure 19. Estimated share of registered unemployed among emigrants with registered labor market
experience, and average last monthly amount received in benefits by emigrants before departure,
2005-2011
In current prices
Source: Calculation with State Social Insurance Agency (SSIA) and State Employment Agency (SEA) data.
Emigrants' age structure is used for assigning weights to individuals (excluding retirees) permanently leaving
both SSIA and SEA datasets in between January 2005 and August 2011 (to allow one year abroad for those who
left most recently).
As discussed in detail in Hazans and Philips (2010), during the growth period, emigration
was not the only cause for the decline of unemployment. Hazans (2013c: Figure 4.18)
demonstrates that increase in job vacancy rate (especially in manufacturing and construction, as
well as for semi-skilled manual workers) outpaced emigration in 2005-2007 (even more so in
2005-2006). By contrast, during the jobless recovery of 2010-2011, job vacancy rate was either
roughly constant at a very low level or growing at a much slower pace than emigration. Moreover,
the fastest growth in job vacancy rate refers to high-skilled non-manual jobs. This is consistent
with an increasing share of university graduates among the emigrants.
0%
10%
20%
30%
40%
50%
,0
,10
,20
,30
,40
,50
2005
2006
2007
2008
2009
2010
2011
Share of registered unemployed
Average last monthly amount of
state and social benefit
Average share of registered unemployed among emigrants (estimate)
Average last monthly amount of state and social benefits received by emigrants (estimate)
39
Several studies have used large macro-econometric models to estimate the effect of
emigration on the rate of unemployment in sending countries, including Latvia; see Holland et al.
(2011) and European Commission (2012, pp. 275-276) for a summary. In particular, Barrel et al.
(2007, Tables 3 and 4) estimate that migration contributed to reduce the rate of unemployment in
Latvia by 2.4 percentage points over the four year period of 2005-2008. Holland et al. (2011),
however, find a much smaller effect. Zasova (2012) developed a model which sets the estimated
contribution of emigration to the decline in the non-accelerating inflation rate of unemployment
(NAIRU) after EU enlargement, at 0.4 points (applying our emigration estimates).
A major focus of public debate in Latvia is the question whether emigration has already led to
labor shortages, as it had in 2005-2007 (see Hazans and Philips, 2010; and Rutkowski, 2007 for
discussion and evidence). While shortages indeed develop in particular occupations (e.g. health
professional, especially in public hospitals outside the capital city, and top-qualified IT
professionals with specific experience in particular sectors), employers and potential investors
complain that despite high unemployment they generally cannot find qualified workers,
suggesting that unemployment in Latvia is largely structural. Survey data, however, provide only
limited support for these claims. The highest proportion of enterprises reporting labor shortages is
found in the construction sector and among large manufacturing firms, but even there it peaks at
about 20% in late 2012, and at any rate remains below 10% in trade and services (Hazans (2013c:
Figure 4.19). A more detailed analysis by Anosova et al. (2012) and Hazans (2013a, 2013b) also
seems to refute the hypothesis that Latvian unemployment is structural (i.e. that available
unemployed are not suited for most of the vacancies offered). Difficulties if finding relevant
employees concern only a small share of businesses and a small proportion of available vacancies.
Nevertheless, labor shortages will inevitably become a serious challenge in future, as that the
cohorts of labor market entrants are expected to be smaller than those of leavers (a situation
exacerbated by emigration, but that would have occurred in any caseIncreased propensity to
emigrate tends to reduce labor supply and make it more elastic, thus increasing real wages and
narrowing the gap between the marginal productivity of labor and pay, but also forcing employers
to lower hiring standards (for a discussion of the latter point, see Hazans and Philips, 2010).
Through real wages, emigration also contributes to increases in consumer prices. At the same
time, however, through falling domestic demand, it also exerts influence in the opposite direction.
Barrel et al. (2007, Tables 3, 4) estimate that over the four year period of 2005-2008
emigration contributed 0.8 percentage points of inflation in Latvia and Lithuania, and 0.2 points in
Estonia. Hazans (2013c: Figure 4.20) applies a macro-econometric model developed by Zasova
(2012) and finds an increase of real wages by 2.5% during the period 2001-2010 compared to a
zero-emigration scenario. Elsner (2013a, 2013b) shows that in Lithuania, a one-percentage-point
increase in emigration rate increases the real wage of men on average by 1%, and that over the
period of five years emigration increased the wages of young workers by 6%, while it had no
effect on the wages of old workers. All these estimates seem quite low. The European
Commission (2012, p. 276) notes that this might be due to aggregation bias and that the effects for
specific skill groups, occupations or sectors might be significantly larger. Furthermore, as noted
by Hazans and Philips (2010), macro-models might underestimate the effect of emigration on real
wages because these models do not account for the monopsonistic structure of the labor market, in
particular the threat of a substantial fall in labor productivity when a firm loses not just a marginal
worker but, say, half of its workforce. Scale effect, work organisation problems, and the inability
to compete for publicly financed projects can all be underlying factors.
40
The overall economic impact of emigration results mainly from a reduction of the labor force.
This effect might be reinforced if emigrants are on average more skilled than non-migrants or
mitigated if they are less skilled. Emigrants' remittances, on the other hand, can partly or fully
compensate the loss of output (in 2013, remittances amounted to 4.5% (respectively, 2.5%; 1.8%)
of Lithuanian (respectively, Latvian, Estonian) GDPxxix but this is unlikely to last forever,
especially when emigration becomes increasingly permanent, as in the case of the Baltic
countriesxxx. For the period of 2004-2009, Holland et al. (2011), assuming a net outflow of just
2.5% of the population estimated the long-term effect on Latvian real GDP to be 3.3%, only half
of which has been compensated by remittances during the same period (European Commission,
2012: 278). Clearly, the overall long-term effect of losing 10% of a country's population, as was
the case for Latvia (see Figure 3) would be much larger. Assuming a production function with the
share of labor about 0.6, the loss of 15% of the labor force implies a permanent reduction of
about 9 percentage points in potential output.xxxi The latter estimate is, however, likely too high
since domestic productivity of at least three quarters of emigrants was below median productivity
of all legally employed persons in Latvia (Hazans 2013c: Figure 4.21).
As shown in Hazans (2008; 2013c), return migrants in Latvia are on average more productive
than otherwie similar non-migrants, but as long as their number is small, this will not be sufficient
to compensate for brain drain.
By reducing population and hence domestic market size, emigration discourages investment –
both foreign and domestic. This is reinforced by the threat of labor shortages (Kugler and
Rapoport, 2005; Javorcik et al., 2011; Gormsen and Pytlikova, 2012). While theoretical
considerations suggest that investment from and trade with countries hosting large numbers of
recent emigrants from Latvia should substantially increase, this is yet to happen, if at all.
However, according to a survey of emigrants from Latvia, 25% of the post-2000 emigrants plan to
start a business in Latvia or to help their employers to establish business relations with Latvia
(Hazans 2015c). Fostering the diaspora’s engagement in economic and social development of the
Baltic countries seems to be the most perspective way to realize gains from emigration. One of the
noteworthy iniatives in this regard is World Latvian Economics and Innovation Forum. About 340
entrepreneurs, engineers, lawyers, scientists and other professionals of Latvian origin from 21
countries took part in the first event of the series in July 2014, and the second one is scheduled for
July 2015.xxxii
10. Conclusion
This chapter offers a theoretical framework and empirical evidence for understanding the
patterns of emigration from the Baltic countries in the 21st century. The focus is on emigrants
selectivity with respect to human capital, ethnicity and citizenship, as well as on labor market
outcomes. We also assess the demographic and economic implications of recent emigration.
During the Great Recession, new emigration wave emerged from each of the three Baltic
countries as a reaction to dramatic rise of unemployment and fall of household income. By 2013,
despite resumed economic growth and massive outflow of population during the previous years,
the Baltic countries featured outflows of nationals well above the pre-crisis levels and high
41
potential for further emigration. The main reasons for it include wide income gap with the old
Europe, the network effect, non-economic push factors (such as disappointment and loss of
perspective - especially in Latvia and Lithuania), weak social protection and insufficient labor
demand in the home countries. Emigration has become “the new normal”, and the Baltic diasporas
will keep growing in the years to come.
While emigration from the Baltics helped to alleviate the economic and social implications
of the crisis by absorbing slack labor and bringing in substantial amount of remittances, in the
longer term it puts at risk reproduction of the populations, sustainability of social security systems
and economic development of these countries. The demographic risk is most pronounced in
Latvia. The countries which host most of the Baltic emigrants (the UK, Ireland, Germany and the
Nordic countries) have better demographic prospects than the Baltics.
Except for the pre-accession period, university graduates were not over-represented among
the Baltic residents working abroad while still attached to their households at home (most of these
mobile workers were medium-educated). However, by 2011, in European OECD countries the
shares of tertiary-educated individuals were substantially higher among (the total stock of) post-
2000 immigrants from Latvia and Lithuania than among similarly aged stayers in these countries.
Brain drain was taking place both as "diploma drain" and via emigration of secondary-educated
youth who received their university degrees in the destination countries and settled there (or
elsewhere abroad). Evidence from the main destination countries suggests that post-accession
Baltic emigrants were less educated than the pre-accession ones - plausibly, because expected
gains were higher for the low- and medium-skilled than for high-skilled, while introduction of the
free movement of labor has diminished the comparative advantage in mobility previously enjoyed
by the better-educated.
Except for emigration from Estonia to Finland, brain drain from the Baltics has intensified
during the crisis (plausibly, reflecting rise of general disappointment and non-economic reasons
for emigration in Latvia and Lithuania). Return migration is unlikely to compensate for the brain
drain, because the share of university graduates among the returnees is either lower or (in some
periods) just slightly higher than among similarly aged stayers.
Ethnic minorities (mostly Russian-speaking) were slightly over-represented among
emigrants from Latvia but under-represented among their Estonian counterparts (likely because
similarity between Estonian and Finnish languages gives ethnic Estonians an advantage over
Russians in Finland's labor market). Both for Estonia and Latvia, the share of ethnic minorities
among mobile workers abroad moved together with the ethnic gaps in employment, declining in
the post-accession period and increasing during the crisis. Moreover, selectivity indexes of
minority individuals with and without Estonian/Latvian citizenship most of the time moved in
opposite directions, reflecting the dual disadvantage faced by residents of Latvia and Estonia
holding either "noncitizen" or Russian (or other non-EU) passports: they have worse labor market
outcomes at home (especially in the bad times) and are not covered by free mobility provisions
within EU. Once emigrated, Russian-speakers are less likely to return than ethnic Latvians or
Estonians.
42
Labor market outcomes of the Baltic emigrants vary across European destinations. In the
UK, which hosts most of Latvian and Lithuanian expats, they fare much better than natives and
than stayers in their home countries; the same is true for Baltic-born youth (but not adults) in
Ireland and Finland. Baltic emigrants with college diplomas enjoy the highest employment rates,
but skill-driven employment gaps are substantially smaller among emigrants than among both
natives in the same countries and stayers.
We find evidence that high-educated emigrants experience downskilling vs. last occupation in
the home country less often than skill underutilization or formal overeducation; in other words,
overeducation is not always caused by emigration. All measures of brain waste are substantially
higher among women. Emigrants in the UK and Ireland are more often subject to brain waste than
elsewhere in EU/EFTA.
Findings from the literature suggest that post-enlargement emigration from the Baltic helped
to contain unemployment but had a negative effect on GDP (and permanent negative effect on
potential output), while wages of the stayers increased compared to zero-emigration scenario. By
reducing population and hence labor force and domestic market size, emigration discourages
investment; however, there is evidence that Fostering the diaspora’s engagement in economic
development of the Baltic countries can partly compensate the negative effects.
REFERENCES
Anniste, K., T. Tammaru, E. Pungas, and T. Paas (2012). Emigration after EU Enlargement: Was
There a Brain Drain Effect in the Case of Estonia? Available at SSRN:
http://ssrn.com/abstract=2070036
Anniste, K. and T. Tammaru (2014). Ethnic differences in integration levels and return migration
intentions: A study of Estonian migrants in Finland. Demographic research 30: 377-412
Anosova, D., K. Sonin, A.Vanags and A. Zasova (2013), Structural or Cyclical? Unemployment
in Latvia Since the 2008-09 Financial Crisis.CEPR Discussion Paper No. DP9525.
Available at SSRN: http://ssrn.com/abstract=2288490
Aptekar, S. (2009), “Contexts of exit in the migration of Russian speakers from the Baltic
countries to Ireland”. Ethnicities 9(4): 507-526
Aujean, L. (2012),Recent trends in Intra-EU Mobility: A Focus on CEE Member States.
Presentation at Advisory Committee on Free movement of Workers”, 30 October 2012.
Blanchflower, D. G. and C. Shadforth (2009), “Fear, unemployment and migration”, Economic
Journal, 119( 535):F136-F182.
Borjas, G. J. (1987), “Self-selection and the Earnings of Immigrants”, American Economic
Review, Vol. 77, No. 4, pp. 531-553.
Borjas, G. J. (1999), “Immigration and Welfare Magnets”, Journal of Labor Economics, Vol. 17,
No. 4, pp. 607-637
43
Broka, A. (2011), “The Social Security of Latvian Families in Age of Economic Migration”, in
B. Zepa and E. Kļave (eds), Latvia. Human Development Report 2010/2011: National
Identity, Mobility and Capability. Riga: University of Latvia Press, pp. 102-106.
Brucker, H., T. Baas, S. Bertoli, T. Boeri et al (2009), “Labor Mobility within the EU in the
Context of Enlargement and the Functioning of the Transitional Arrangements, Nurember g:
European Integration Consortium (IAB, CMR, fRDB, GEP, WIFO, wiiw).
Department for Work and Pensions (2014). National Insurance Number Allocations to Adult
Overseas Nationals Entering the UK.
Elsner, B. (2013a). Does emigration benefit the stayers? Evidence from EU enlargement. Journal
of Population Economics, 26 (2): 531-553
Elsner, B. (2013b). Emigration and wages: The EU enlargement experiment. Journal of
International Economics, 91 (1): 154–163
European Commission (2010), Employment in Europe 2010, Publications Office of the European
Union, Luxembourg,
http://ec.europa.eu/social/main.jsp?catId=738&langId=en&pubId=593.
European Commission (2011), Labor Market Developments in Europe, 2011,
http://ec.europa.eu/economy_finance/publications/publication_summary-20110809_en.htm.
European Commission (2012). Employment and Social Developments in Europe 2011.
European Commission (2013). EU Employment and Social Situation. Quarterly Review - June
2013. Luxembourg: Publications Office of the European Union.
Gormsen, C. and M. Pytlikova (2012), “Diasporas and Foreign Direct Investments”, Paper
presented at EALE Annual Conference, Bonn, 20-23 September 2012,
www.eale.nl/Conference2012/program/Parallel%20session%20B.htm.
Hazans, M. (2003), “Potential Emigration of Latvian Labor Force after Joining the EU and its
Impact on Latvian Labor Market”, available at SSRN, http://ssrn.com/abstract=739305.
Hazans, M. (2007), “Coping with Growth and Emigration: Latvian Labor Market Before and
After EU Accession”, available at SSRN, http://ssrn.com/abstract=9711908
Hazans, M. (2008), “Post-enlargement Return Migrants’ Earnings Premium: Evidence from
Latvia”, EALE 2008 Paper, No. 541, available at SSRN, http://ssrn.com/abstract=1269728.
Hazans, M. (2010), “Ethnic Minorities in Latvian Labor Market, 1997-2009: Outcomes,
Integration Drivers and Barriers”, in N. Muižnieks (ed.), How Integrated is Latvian Society?
An Audit of Achievements, Failures and Challenges, Riga: University of Latvia Press,
pp. 125-158.
Hazans, M. (2011a), “Labor Market Integration of Ethnic Minorities in Latvia”, in M. Kahanec
and K.F. Zimmermann (eds), Ethnic Diversity in European Labor Markets: Challenges and
Solutions, Cheltenham, UK Northampton, MA, USA: Edward Elgar, pp. 163-197.
Hazans, M. (2011b), “The Changing face of Latvian Emigration, 2000-2010”, in B. Zepa and
E. Kļave (eds), Latvia. Human Development Report 2010/2011: National Identity, Mobility
and Capability. Riga: University of Latvia Press, pp. 77-101.
Hazans, M. (2012), “Selectivity of Migrants from Baltic Countries Before and After Enlargement
and Responses to the Crisis”, in B. Galgóczi, J. Leschke and A. Watt (eds), Intra-EU
Migration in Troubled Times: Skills Mismatch, Return Migration and Policy Responses,
Farnham, UK: Ashgate, pp. 169-207.
Hazans, M (2013a), “Structural or cyclical? Unemployment in Latvia since the 2008-2009
Financial Crisis”. Public presentation, March 1, 2013, http://www.sseriga.edu/en/news-and-
events/upcoming-events/event-archive/2013/presentation.html
Hazans, M. (2013b), “Developments in the Latvian Labor Market”. Presentation at the conference
44
Beyond Recovery: Foundations for Inclusive Growth in Latvia, Riga, Latvia, June 3-4 2013,
http://www.worldbank.org/content/dam/Worldbank/Feature%20Story/ECA/Poland%20prese
ntations%20merged.pdf
Hazans, M. (2013c). “Emigration from Latvia: Recent trends and economic impact”, in
OECD (2013), Coping with Emigration in Baltic and East European Countries, OECD
Publishing, pp. 65-110. doi: 10.1787/9789264204928-en
Hazans, M. (2014a). “Emigration Intentions and Fertility Potential in Latvia”. Presentation at the
conference Labor Mobility and Transnationalism in the Nordic-Baltic Region, Tallinn,
Estonia, March 7, 2014.
Hazans, M. (2014b). Migration experience of the Baltic countries during and after the economic
crisis". Presentation at the conference Coping with Emigration in the Baltic and East
European Countries: Lessons Learned and New Challenges, Riga, Latvia, April 2, 2014,
http://www.sseriga.edu/en/news-and-events/upcoming-events/copingwithemigration.html
Hazans, M. (2015a). "Return Migration Intentions to Latvia, Based on Recent Survey of
Emigrants". Presentation at the conference Migration in the Nordic - Baltic Region. New
Trends of Labour Migration - Ready for the Changes? Tallinn, Estonia, March 27, 2015,
http://www.norden.ee/images/regionaal/info/migra/conference_27032015/Hazans_migra2015.pdf
Hazans, M. (2015b). “Poland and the Baltics: The nature of the four emigration waves over 2000-
2013”. Peresentation at The World Bank—European Commission Joint Seminar on Active
Aging, Brussels, Belgium, 11 March 2015,
http://www.age-platform.eu/images/stories/PPT/Hazans_Brussels_11March2015.pdf
Hazans, M. (2015c). “For many, emigration from Latvia is a success story” ("Для многих отъезд
из Латвии история успеха”, in Russian). Открытый город, 5/2015, 52-57. Available at
http://www.freecity.lv/obshestvo/23637/
Hazans, M. and K. Philips (2010), “The Post-Enlargement Migration Experience in the Baltic
Labor Markets”, in M. Kahanec and K.F. Zimmermann (eds), EU Labor Markets After Post-
Enlargement Migration, BerlinHeidelberg: Springer, pp. 255-304, also available as IZA
Discussion Paper No. 5878, http://ftp.iza.org/dp5878.pdf.
Holland, D., T. Fic, A. Rincon-Aznar, L. Stokes and P. Paluchowski (2011), “Labor Mobility
within the EU – The Impact of Enlargement and the Functioning of the Transitional
Arrangements”, Study commissioned by the Directorate-General for Employment, Social
Affairs, and Equal Opportunities, NIESR, London.
http://ec.europa.eu/social/BlobServlet?docId=7120&langId=en
Hughes, J. (2005), ‘Exit’ in Deeply Divided Societies: Regimes of Discrimination in Estonia
and Latvia and the Potential for Russophone Migration”, Journal of Common Market
Studies,43 (4): 739-762.
Ivlevs, A. (2013). Minorities on the move? Assessing post-enlargement emigration intentions of
Latvia‘s Russian speaking minority. The Annals of Regional Science 51(1): 33-52.
Javorcik, B.S., Ç. Özden, M. Spatareanu and C. Neagu (2011), “Migrant Networks and Foreign
Direct Investment”, Journal of Development Economics, Vol. 94, pp. 231-241”.
Kaczmarczyk, P., (2013), Matching the skills of return migrants to labor market needs in Poland,
in OECD (2013), Coping with Emigration in Baltic and East European Countries, OECD
Publishing, pp. 111-126. doi: 10.1787/9789264204928-en
Kaczmarczyk, P., M. Mioduszewska and A. Żylicz (2010), “Impact of the Post-accession
Migration on the Polish Labor Market”, in M. Kahanec and K.F. Zimmermann (eds), EU
Labor Markets After Post-Enlargement Migration, Springer, Berlin–Heidelberg, pp. 219-
253
45
Kahanec, M. (2012), "Labor Mobility in an Enlarged European Union" In: A. F. Constant, K. F.
Zimmermann (eds.), International Handbook on the Economics of Migration, Edward Elgar
2013, Cheltenham, UK, and Northampton, pp. 137-152
Kahanec, M. and B. Fabo (2013), "Migration Strategies of the Crisis-Stricken Youth in an
Enlarged European Union", Transfer: European Review of Labour and Research. 19 (3):
365-380
Kahanec, M., M. Pytlikova and K. F. Zimmermann (2014), "The Free Movement of Workers in
an Enlarged European Union: Institutional Underpinnings of Economic Adjustment", IZA
Discussion Paper No. 8456
Kahanec, M. and K.F. Zimmermann, eds (2010), EU Labor Markets After Post-Enlargement
Migration, Berlin–Heidelberg: Springer.
Kugler, M. and H. Rapoport (2005), “Skilled Emgration, Business Networks and Foreign Direct
Investment”, CESifo Working Paper Series, No. 1455.
Masso, J. and K. Krillo (2011), "Labour markets in the Baltic States during the crisis 2008-2009:
the effect on different labour market groups" The University of Tartu Faculty of Economics
and Business Administration Working Paper No. 79 . http://dx.doi.org/10.2139/ssrn.1734378
Masso, J., Eamets, R. & Motsmees, P. (2014). Temporary migrants and occupational mobility:
evidence from the case of Estonia .International Journal of Manpower, 35 (6): 753-775.
Mierina, I. (2015), "Return migration to Latvia: Problems, policies, perceptions and perspectives",
Presentation at the conference How to Improve Intra-European Mobility and Circular
Migration? Fostering Diaspora Engagement. Riga, Latvia, 11-12 May 2015,
https://eu2015.lv/images/Kalendars/MFA/Inta_Mierina_Presentation_Riga.pdf
OECD (2012), Connecting with Emigrants. A Global Profile of Diasporas, OECD Publishing,
Paris, http://dx.doi.org/10.1787/9789264177949-en .
OECD (2014a), International Migration Outlook 2014, OECD Publishing.
http://dx.doi.org/10.1787/migr_outlook-2014-en
OECD (2014b), “Database on Immigrants in OECD Countries (DIOC) 2010/11”
Praakli, K. (2011). "The new Estonian community in Finland". In: Grunthal, R. and M. Kovacs
(eds). Ethnic and linguistic context of identity. Finno-ugric minorities.Helsinki: University
of Helsink: 217-247.
Pungas, E., O. Toomet, T. Tammaru and K. Anniste (2012). "Are better educated migrants
returning? Evidence from multi-dimensional education data", NORFACE Migration
Discussion Paper No. 2012-18.
Rose, R. (2000), “New Baltic Barometer IV: A Survey Study”, Centre for the Study of Public
Policy, University of Strathclyde, Glasgow.
Rutkowski, J. (2007), “From the Shortage of Jobs to the Shortage of Skilled Workers: Labor
Markets in the EU New Member States”, IZA Discussion Paper, No. 3202, Bonn.
Saukienė, I. (2011). Poll: Half of the population somehow thinks about emigration (in
Lithuanian). http://www.delfi.lt/news/daily/lithuania/apklausa-apie-emigracija-vienaip-ar-
kitaip-masto-puse-salies-gyventoju.d?id=51966417
Samoškaitė, E. (2012). Poll: Many Lithuanians wish to emigrate permanently [In Lithuanian]
http://www.delfi.lt/news/daily/emigrants/apklausa-daugelis-lietuviu-emigruoti-visam-laikui-
netroksta.d?id=56287685
Sipavičiene, A. and V. Stankuniene (2013). The social and economic impact of emigration on
Lithuania, in OECD (2013), Coping with Emigration in Baltic and East European
Countries, OECD Publishing, pp. 65-110. doi: 10.1787/9789264204928-en
46
Sjaastad, L.A. (1962), “The Costs and Returns of Human Migration”, Journal of Political
Economy, Vol. 70, No. 5, pp. 80-93.
Tammaru, T. and T.Kulu (2003). "The ethnic minorities of Estonia: changing size, location and
composition. Eurasian Geography and Economics 44(2): 105-120
Tarum, H. (2014) „Migration Potential of Working-age Population in Estonia in 2013", Tallinn:
Policy Analysis, Series of the Ministry of Social Affairs No 2/2014 eng.
Veidemann, B. (2010). „Migration Potential of Working-age Population in Estonia in 2013",
Tallinn: Policy Analysis, Series of the Ministry of Social Affairs No 8/2010 eng.
Zaiceva, A. and K. F. Zimmermann (2008). "Scale, Diversity, and Determinants of Labour
Migration in Europe." Oxford Review of Economic Policy, 24 (3), 428-452
Zaiceva, A. and K. F. Zimmermann (2012). "Returning Home at Times of Trouble? Return Migration of
EU Enlargement Migrants during the Crisis". IZA Discussion Paper No 7111
Zasova, A. (2012), “Econometric Assessment of Performance of Latvian Labor Market”, PhD
Thesis, University of Latvia, Riga.
Zepa, B. and E. Kļave (eds) (2011), Latvia. Human Development Report 2010/2011: National
Identity, Mobility and Capability, Riga: University of Latvia Press.
47
NOTES
i Kahanec and Zimmermann (2010) review evidence on the early post-enlargement mobility in a broader EU context.
Kahanec (2012) extends analysis to cover also the 2007 enalrgement and the beginning of the economic crisis.
Kahanec et al (2014) provide econometric analysis of migration flows in EU15+ EFTA +EU12 over the period of
1995-2010.
ii This was the case both in fixed prices (as shown in Figure 1) and in PPS (not shown).
iii In the case of Estonia, however, the situation might be affected by the fact the main destination country, Finland is
geographically and linguistically very close, which makes psychic cost of staying abroad lower and pressure to
return smaller.
iv NINO statistics reflects only immigrants aged 16 and more; it has been adjusted upwards assuming, for each year
and sending country, the same proportion of children among immigrants to the UK as among immigrants to other
EU/EFTA countries.
v These results are based on data covering outflows to the UK, Germany, the Nordic countries, Ireland, Austria and
the Netherlands, which accounted for more than 90% of the total emigration from each of the Baltic countries to
the EU and OECD destinations in 2012.
vi The official Lithuanian figure for 2010 is an exception: many of the previous years' emigrants signed out from the
population register in 2010 to avoid compulsory health insurance payments, see OECD ( 2012: p. 248).
vii Hazans (2003, Tables A4.1-A4.4) provides a detailed comparison of earnings.
viii These results based on Eurobarometer 64.1 data refer to population aged 18 to 65 years; see Hazans (2012: Table
3) for Estonia and Latvia.
ix Brucker et al. (2009, Tables 6.7- 6.8) in the case of UK in 2004-2007 report returns of just 2% per year of
schooling and finds that 82% of tertiary-educated immigrants from the NMS were over-qualified for their jobs. In the
same period, 40% to 60% of tertiary-educated Estonian and Latvian mobile workers and more than 60% of their
Lithuanian counterparts were over-qualified, according to Hazans and Philips (2010, Figure 7).
x See Figures 1 and 6 above; on wage cuts, see Hazans (2013c: Figure 4.7), Masso and Krillo (2011, Table 14).
xi During the crisis years in Latvia, for workers with less than 20 years of contribution this was the case already after 6
months of registered unmeployment.
xii Hazans (2012, Figure 6.3; 2013c: Table 4.5) provides evidence for Estonia and Latvia
xiii Kaczmarczyk et al. (2010) use SI = GM /GS 1 with similar properties; the advantage of our measure is in having
symmetric (opposite) values for GM /GS = k and GM /GS = 1/k.
xiv For Estonia, this finding is supported also by results in Anniste et al. (2012) who used data on registered
emigration.
xv Estonian case is not perfectly comparable to the other two because some of the Estonians working in Finland
commute to/from Estonia (mostly on weekly basis but in some cases more often); yet substantial part of these
commuters are registered as residents of Finland. This makes exact identification of Estonian emigrants in Finland
difficult. Commuters are, on average, less educated than settled emigrants.
xvi The UK Population Census data are likely to overestimate the share of tertiary-educated immigrants and under-
estimate the share of low-educated. First, the low-educated immigrants are more likely to avoid participation due
to language problems. Second, due to complicated design of the question on educational attainment and specific
terminology used in it, many of the immigrants (especially the low-educaated) choose only the answer "Foreign
qualification". The imputation algorithm used by ONS when producing publicly available tables on educational
attainment by country of birth is not sending-country-specific and classifies 69% of these immigrants as highly-
educated. We assume that the bias is not big enough to change the conclusions by Figure 8 (lower panel)
qualitatively. This assumption is supported from data from online survey of Latvian emigrants conducted in 2014
(see Note xviii below), with non-weighted share of tertiary-educated among 3091 post-2000 emigrants living in the
UK being 40%. However, this issue calls for further investigation.
48
xvii Note that due to data limitations pre- and post-accession are proxied by 2001-2005 and 2006-2009 in Fgure 8
(lower panel) and Figure 9.
xviii The survey has been designed and conducted in the framework of interdisciplinary research project “The
emigrant communities of Latvia: National identity, transnational relations, and diaspora politics“ implemented by
Institute of Philosophy and Sociology, University of Latvia in cooperation with Faculty of Economics and
Management, University of Latvia and supported by European Social Fund Project
2013/0055/1DP/1.1.1.2.0/13/APIA/VIAA/040.
Comparison of the respondents' distribution by host country, age, gender and period of leaving with data from other
sources does not show any significant selection bias. Moreover, distribution of repondents from the UK and Ireland
by educational attainment is largely consistent with the data from the yeat 2011 Population Censuses in these
countries.
xix Except for the most recent emigrants, the stock SI are smaller that the flow ones because during the period between
arriving to the host country and 2014, the share of university graduates among emigrants was growing slower than
among young stayers.
xx Anniste et al. (2012: Table 1), using data on registered emigrants from Estonia, find that the proportion of
minorities among emigrants declined from 48% in 2000-2003 to 28% in the post-accession period (2004-2008).
This also support our expectations on the pattern of ethnic selectivity.
xxi See Kahanec and Fabo (2013) for a recent analysis of gender and family effects on emigration intentions in the EU.
xxii Ministry of Education and Science, unpublished data
xxiii Due to data limitations, Germany, Austria an Switzerland are excluded.
xxiv See endnote xviii for details.
xxv Note that median, across all years, tenure of Latvian mobile workers in the LFS data is less than 1 year.
xxvi Eamets et al (2013) provide evidence for substantial downskilling among Estonians working abroad.
xxvii Note that in the post-accession period, the total inflow or returnees during the moving 10-year window was
growing over time.
xxviii By contrast, Eamets et al (2013) do not find an earnings premium for return migrants in Estonia.
xxix World Bank online database, assessed on June 30, 2015.
xxx In 2014, only 35% of Latvian emigrants sent home remittances on a regular basis, according to the survey
described in endnote xviii and Hazans (2015a, 2015c).
xxxi The effect of emigration on total labor force participation is theoretically ambiguous. Changes in the age structure
caused by emigration suggest a negative effect, while higher real wages and lower hiring standards tend to increase
the participation rate, especially among disadvantaged groups (Hazans and Philips, 2010; Hazans, 2011a). In fact, the
activity rate of the Latvian working age population was much higher in 2011-2012 than in the pre-accession period,
but it could have been even higher in absence of emigration.
xxxii For details, see http://www.ieguldilatvija.lv/
... The time dimension is central in studying highly skilled migration from Eastern Europe. Research shows a more pronounced pattern of emigration of highly skilled before Latvia joined the EU in 2004 than during post-accession years (Hazans 2016). The Great Recession of 2009-2010 changed the profile of emigrants. ...
... The Great Recession of 2009-2010 changed the profile of emigrants. During the years of this economic crisis and its aftermath, net emigration outflow from Latvia exceeded the number of people who had left Latvia between 2000 and 2008 (Hazans 2016). In addition, the most significant increase was among highly educated emigrants, with non-economic reasons driving it. ...
... In addition, the most significant increase was among highly educated emigrants, with non-economic reasons driving it. Hazans (2016) concluded that by 2011, university graduates were over-represented among post-2000 Latvian emigrants in European OECD countries; in main non-European destinations they held higher educational credentials than those in Europe (Hazans 2019). ...
Article
This paper empirically examines whether former international students, i.e., global graduates, are more likely to be ‘liquid migrants’ than those who hold higher education degrees from home and the factors which enable or inhibit transnational mobility among global graduates. Based on an analysis of a subsample from a large-scale web survey of Latvian migrants, this paper finds that highly-skilled Latvian and global graduates represent two different groups of migrants. On the one hand Latvian graduates appear to migrate due to employment or family reasons and seek stability when doing so, corresponding more closely to the profile of sedentary migrants. Former international students from Latvia, on the other hand, are more likely to explore different destinations, even though most of them have lived in no more than two countries, excluding their country of origin. Global graduates who are older, male, hold a doctoral degree, are more recent emigrants, and live outside Europe and the US, are more likely to engage in a ‘liquid migration’ lifestyle. The empirical analysis and findings in this paper suggest alignment with the conceptualisation of ‘liquid migration.’
... Moreover, some evidence points to a contradictory pattern compared to Castles' (2009) remarks. For example, Hazans (2016) showed how three Baltic countries increased their emigration flows after the 2008 crisis. These flows were mainly driven by other European Union members and sustained by the Schengen mobility agreement (Hazans 2016). ...
... For example, Hazans (2016) showed how three Baltic countries increased their emigration flows after the 2008 crisis. These flows were mainly driven by other European Union members and sustained by the Schengen mobility agreement (Hazans 2016). Similarly, between the years of 2008 and 2012, nations such as Australia, Norway, Austria, and Finland, far from reducing their immigration flows, reported an increase of these types of foreigners (Organisation for Economic Co-operation and Development (OECD), 2014). ...
Article
Economic and social conditions have deteriorated worldwide during the COVID-19 pandemic. Migration theory and international organizations indicate that these increasingly fragile social conditions represent powerful incentives to migrate. Normally, studies addressing international migration and COVID-19 focus on transit and destination countries, with substantially less literature centered on origin nations. Trying to close that gap, the present article aims to identify and quantify economic determinants that explain the intention of Salvadorians to migrate abroad. Using a probabilistic sample and a logistic model, a number of renowned economic variables for migration studies were used to investigate Salvadorian’s intention to emigrate. Results demonstrated a stark reduction in migration intentions in 2020. Moreover, the risk of losing one’s job is by far the most prominent factor explaining the intention to migrate. Other aspects, such as employment and salaries, also showed statistically significant values. Additionally, results report women being less likely to migrate and age to have a negligible effect. The text concludes by indicating some public initiatives that could be implemented to support people who choose to act upon their intentions and embark on emigration.
... Meanwhile, Rogoff and Reinhart's work on austerity was found deficient as a result of the inclusion of significant spreadsheet errors (Herndon, Ash, and Pollin 2013). Some argue that austerity was accepted in Latvia because the ideas found a partial resonance in Latvian culture (Ozolina 2019), because ethnic political tensions superseded economic ones and as a result of the significant exodus of people (Hazans 2016) that served to dilute political pressures against austerity (Sommers, Woolfson, and Juska 2014). Heterodox critiques of Latvia's development model have been produced, albeit with varying areas of focus (Berzins 2014;Weisbrot and Ray 2011;. ...
... The demographic implications of this restoration through austerity were catastrophic. Young people exited Latvia in the hundreds of thousands, which risked turning the country into a retirement home and nature preserve (Hazans, 2016). Latvia's population peak was reached in 1989 at nearly 2.7 million people, surpassing its previous record of 2.5 million people at the onset of WW I in 1914(CEIC Data, 2017. ...
Article
Full-text available
This article enumerates several sectors where Latvia’s economic performance can be enhanced. Yet, it also delivers by way of a long introduction an overview of Latvia’s economic development in the post-Soviet period from a heterodox perspective that hints at why good options might not have been selected in the past. Following the above overview, we provide a detailed sectoral analysis of Latvia’s economy since the 2008 financial shock that reveals continued hindrances to its development. We conclude by outlining development-friendly tax and industrial policy recommendations to promote Latvia’s continued transformation into a more modern, sustainable, socially responsible and digitally enabled economy.
... On the other hand, mass emigration of skilled and younger working-age population can lead to skill depletion in the local labour market, reducing productivity and undermining Lithuania's long-term development (Thaut, 2009). According to Statistics Lithuania (2021), in 2001-2020, emigration consisted mainly of primary-working-age individuals, and it is estimated that approximately 20-30% of the working-age emigrants were highly skilled (Hazans, 2016;Karalevičienė & Matuzevičiūtė, 2009). The pessimistic position is also supported by the fact that one third of recent graduates chose emigration in the context of low birth rates and sharply declining university enrolment numbers (OECD, 2018). ...
Article
Full-text available
This study explores the economic impact of population emigration with special reference to the case of Lithuania. For this reason, we developed a SVAR model and applied related IRF and FEVD tools using quarterly data for the period of 2001-2020. Our findings reveal that a positive shock in emigration is related to lower unemployment. It is also found that the increased emigration is linked to higher real wage growth but with a lower confidence interval. Moreover, our estimates suggest that international out-migration increases real GDP growth in the short term, with no significant effects in the long run perspective. Finally, we found that most of the emigrants-to-be were inactive for a long term prior to departure, which offers a new look into the consequences of Lithuanian emigration, suggesting that the economic losses of emigration could be overstated. This study contributes to the knowledge about the impact of emigration on the economy and specifies directions for further studies in the field.
... In the decade after the collapse of the Soviet Union, migration of all minority groups was high, Russians included (18%, according to the Census). Hazans (2016) finds that Russian-speaking population in pre-EU accession time had lower labour market outcomes and hence contained greater emigration potential which materialised in overrepresentation of Russian-speakers among emigrants. While in the 2000s migration of ethnic Russians was double that of Latvians (-11.8%); between 2011-16 the numbers almost converged: the difference was only 1.7 percdentage points (Hazans, 2019). ...
Article
Data on migration flows suggest that young people are highly mobile. Yet, there are gaps in the evidence concerning the factors driving young people’s international migration in Latvia. Latvia is a potentially interesting case because of the high rate of migration from the country, but also because it is a complex ethnic and linguistic environment. Latvian and Russian speaking populations are shown to have diverse migration drivers, and this study addresses the differences in attitudes to studying and living abroad for adolescents at the time of school graduation. Literature suggests that Russian-speaking population have higher propensity to emigrate. We aim to contribute to the literature by exploring differences in migration intentions between Latvian and Russian speaking high school graduates.This study is based on individual-level survey data of secondary school graduates in Latvia in 2020, amidst COVID-19 pandemic. We analyse the strength of migration intentions from definitely not leaving Latvia to surely planning to emigrate. We conclude that Russian-speakers exhibit stronger intentions to emigrate compared Latvian-speaking youngsters, driven by wider networks and expected higher returns to their human capital abroad.
... The economic crisis was very deep in the Baltic states and this may have strengthened the emigration incentives. Surveys of the motives of emigrants from the Baltic states suggest that outward migration was a strategy for coping with low incomes, high unemployment and limited welfare payments, and also with a perceived lack of prospects for the future (Hazans 2016). Moreover, the countries are small and culturally open, and so the younger generations in the Baltic states may have felt comfortable moving to another country. ...
Chapter
This chapter discusses key macroeconomic trends in the Baltic states before and after their accession to the European Union in 2004. The countries have seen rapid economic growth, and the income gap with the EU countries in Western Europe has narrowed substantially. The convergence process has however been accompanied by cyclical volatility, and the setback after the global financial crisis was particularly severe. The cooperation of the Baltic states with the EU began in the early 1990s and strengthened over time, so the accession in 2004 did not cause any abrupt changes. Membership however opened for increased economic support from the EU and further liberalisation of trade, capital flows and movements of labour. A key challenge for the Baltic states is to achieve steady economic growth and continued convergence without jeopardising their macroeconomic and financial stability. This calls for judicious domestic policymaking in order that the countries may reap the benefits from continued integration and cooperation with the EU.
Article
The article presents the results of the study of the main trends of the labor market in Central and Eastern Europe during 2010-2021; it also investigates selected labour market outcomes affecting migration processes in Bulgaria, Czech Republic, Estonia, Latvia, Lithuania, Romania, Poland, Slovak Republic, Slovenia, and Hungary. Using econometric modeling, the impact of GDP per capita, annual net earnings per employee, job vacancy rate, unemployment rate on the number of emigrants from these CEE countries was estimated. It is followed by the analysis which controls for dependence of the number of immigrants to CEE countries on labor market outcomes and GDP per capita. The simulation results demonstrate a direct impact of the indicators of net earnings per worker, the level of unemployment and quantity of emigrants from CEE countries. At the same time, there is a direct relationship between job vacancy rate, net earnings and immigration into Bulgaria, Czech Republic, Estonia, Latvia, Lithuania, Romania, Poland, Slovak Republic, Slovenia, and Hungary.
Article
Full-text available
Emigration countries such as Latvia institutionally should address the negative consequences of the ageing populations' economic and demographic challenges. To this aim, in 2018 network of five regional return migration coordinators was established to foster and nationally encourage return migration. The primary function of the regional coordinator is to foster and facilitate requests of potential return migrants.Drawing on interview materials collected in 2020 with return migration coordinators at the regional level – Latgale, Vidzeme, Zemgale, Kurzeme and Riga, we illustrate good-practice of return migration institutionalisation in Latvia.Our study focuses on three main fields of interest within the work of regional return coordinators: 1) what are the typical return migration paths from the coordinators' perspective? 2) what are the main challenges in coordinators' work? 3) how education-related issues for return migrants are solved?The research reveals that people typically return for family reasons and long for their homeland, traditions, and traditional settings. The r