Society and Economy

DOI: 10.1556/204.2019.010

THE ROLE OF LOYALTY IN MIGRATION

INTENTIONS: THEORY AND EVIDENCE
FROM THE EU*

JÓZSEF GOLOVICS

Department of Comparative and Institutional Economics,

Corvinus University of Budapest, Hungary
E-mail: jozsef.golovics@uni-corvinus.hu

Free movement of persons is one of the fundamental values and achievements of the European
Union, however, intentions towards mobility vary across and within the member states. Economic
literature has remarkable theories to explain migration flows and individual selection factors of
potential migrants, but it ignores major achievements of other social sciences. This paper builds an
economic framework to incorporate the Hirschmanian concept of loyalty into the microeconomic
(human capital) model of international migration by using interdependent preferences. Hirschman
assumes that even after exiting, loyal people care about their previous communities, thus it imposes
a certain psychological ‘exit tax’ on them. Based on this concept, it is hypothesized that people
with altruistic motives have weaker intentions to migrate, so the presence of loyalty towards others
makes international migration less likely, conveying that loyalty towards local or national com-
munity may be responsible for moderate labor mobility among EU member states. Results show
that attachment to one’s country makes one’s intention to move abroad in the near future less likely,
while loyalty towards one’s city has more moderate impact on their intentions.

Keywords: migration, loyalty, altruism, EU

JEL-codes: F22, D00, D90

* This paper has been supported by the European Union and Hungary and co-financed by the
European Social Fund through the project EFOP-3.6.2-16-2017-00017, entitled “Sustainable,
intelligent and inclusive regional and city models.”

1588-9726 © 2019 The Author

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

173 JÓZSEF GOLOVICS

1. INTRODUCTION

Nowadays, international migration is an important and current issue both at a

European and a global level. According to the UN (2017) there were 258 million
international migrants worldwide in 2016. However, migration decisions in the
21st century are characterized by a radically different environment than those in
previous decades. Thanks to immense technological progress, the costs of mobil-
ity have dropped dramatically (Hoffmann 2010), and this was not accompanied
by a similar decrease in international development disparities.
This environment influences potential migrants. Although various schools of
the migration literature explain causes and main driving forces of migration dif-
ferently (see Massey et al. 1993), none of them deny the importance of interna-
tional disparities and the income gap, as well as the influential role of mobility
costs. However, this implies that circumstances outlined above make migration a
viable alternative for more and more people living in less developed countries.
The European Union is a spectacular example for this, inasmuch as achieve-
ments of the integration, regulation and other facilitation guarantee the freedom
of movement for persons, which undoubtedly lessened mobility costs. Besides
that, possibilities created by technological progress are also available. At the
same time, development disparities are harshly present among old and new mem-
ber states. As an example, country level data show an up to eight-fold difference
in annual net earnings1 counted in euros and a two to three times gap can be seen
even at purchasing power parity (Eurostat 2017).
Therefore, according to neoclassical economics – which predicts the flow of
labor until the international equilibration of wages2 – intense waves of migration
should have been witnessed in this relatively ‘frictionless’ institutional environ-
ment. However, despite any intentions or attempts of the European Commission
(Andor 2014), intra-EU mobility is low. As an example for the puzzling phenom-
enon of this ‘missing migration’, only 3.0 percentage of European citizens lived in
a member state other than their country of citizenship in 2016 (Eurostat 2018).
One cannot rule out the assumption that existing mobility costs are able to
offset benefits of migration despite their extremely low level, however, anoth-
er important factor remains unexplained. Although people typically move from
Central East European (CEE) member states to the Western part of the EU, the
extent of emigration affects CEE countries in a different way. Moreover, migra-

1
Disparities observed in annual net earnings are not exclusive indicators of differences in
standard of living, but they correlate strongly with other relevant factors.
2
Although assumptions of these neoclassical models are not realistic, the EU can serve as the
closest real-life example with its relatively ‘frictionless’ institutional framework.

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

 THE ROLE OF LOYALTY IN MIGRATION INTENTIONS 174

tion rates did not correlate with net annual earnings within this group between
2008 and 2013.
All the above-mentioned facts strengthen the empirical puzzle that develop-
ment and earning disparities might be offset by other factors in potential mi-
grants’ cognitive framework. This paper hypothesizes that emotional ties, namely
the presence of loyalty, is an important factor.
Loyalty is represented through psychic costs of migration in this paper. Men-
tioning the role of psychological costs of migration is not a novel idea in migra-
tion literature. Sjaastad (1962), pioneer of the human capital migration theory,
also discusses the concept, however, he does not treat it as a real resource cost.
Based on the idea that it is hardly measurable, he does not consider psychologi-
cal costs as analitically usable components among the costs of migration. Other
prominent scholars (e.g. Bauer – Zimmermann 1995; DaVanzo 1981; Massey
et al. 1993) simply avoid detailed discussion of this factor, however, they name
it as an important restraining force. Nonetheless, we argue that consideration of
psychological costs can help to discover and understand the causes of the above-
mentioned ‘missing migration’ in today’s changed migrational environment. In
this regard, psychological aspects of migration are not treated as only a type of
mobility cost. Instead, interdependent preferences and Hirschman’s (1970) loy-
alty concept are used to incorporate these psychological ties into the microeco-
nomic model of migration.
In order to test the role of loyalty empirically, we use individual Eurobarom-
eter data on migration intentions from European Union member states. Loyalty is
operationalized as a person’s attachment to their city and country. We expect that
beyond traditional selection factors (e.g. age, gender, family status, labor market
status, etc.) loyalty ceteris paribus significantly decreases the probability of the
intention to migrate. Similarly to previous papers (e.g. Fouarge – Ester 2008;
Sik – Szeitl 2016; Zaiceva – Zimmermann 2008a) logit models are used in the
estimation.
The paper is structured as follows. After the introductory section, we review
the basic microeconomic background of migration in order to situate our theory
in the existing literature (Section 2), then we build our theoretical model of mi-
gration considering loyalty (Section 3). After that, we present the dataset and
methodology of our empirical test (Section 4). After demonstrating and discuss-
ing our results (Section 5), the paper comes to a conclusion.

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

175 JÓZSEF GOLOVICS

2. THEORETICAL BACKGROUND

This paper basically follows the tradition of the neoclassical microeconomic (or
human capital) approach to migration (for further schools of migration theory see
Massey et al. 1993). Thus, our unit of analysis is the individual and migration is
viewed as an individual decision. As Becker (1962) emphasizes, people seek to
economize rationally with their human capital: one might invest in it to gain max-
imal return. Based on Sjaastad’s (1962) work, also theorized by Hart (1975) and
Todaro and Maruszko (1987), migration is also a form of human capital invest-
ment. It has costs, but migrants are willing to undertake them in the hope of future
benefits, which might lead to a better life in a foreign country. This investment
decision is considered to be self-interested and preceded by rational cost-benefit
calculation. Thus, potential migrants choose to leave their country of origin if
benefits overwhelm costs of migration, and therefore it results in a net increase of
benefits. In this calculation they consider international wage differences, stand-
ard of living disparities and labor market conditions. These market-type factors
can be supplemented by further aspects that represent the governments’ involve-
ment in real economy (e.g. amount of taxes or public spending on unemployment
benefits and other welfare expenditures). The net value of the latter is hereinafter
referred to as net public service level.
The significant impact of the abovementioned factors on migration is verified
in several empirical studies by using various methods at different levels. As for
benefits, Ahman et al. (2008) proved the macro-level relevance of real wages and
unemployment by using time-series from Pakistan. By using panel macro data
from OECD countries, Mayda (2010) analyzed bilateral flows and found that
wages and development (proxied by GDP) have a significant effect on migration.
Sprenger (2013) analyzed intra-EU migration with panel methods and stressed
the causal relationship between real economy conditions (per capita GDP, unem-
ployment) and migration. Geis et al. (2008) focused on four developed countries
and used multinomial logistic regression on micro data in order to prove that
factors defining the net public service level in receiving countries also play an
important role in migration decisions. Kureková (2013) used quantitative, as well
as qualitative methods and found that the net public service level and welfare
systems in countries of origin also affect intra-EU migration.
Benefits of migration may be balanced by its costs. These mobility costs con-
sist of direct costs of travelling, moving and gathering information about job and
accommodation opportunities as well as indirect costs such as opportunity cost
of the latter actions or burdens of dealing with regulation (e.g. administrative du-
ties imposed by immigration or visa policy), institutional, cultural and language
differences.

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

 THE ROLE OF LOYALTY IN MIGRATION INTENTIONS 176

The significance of these costs on the migration decision was also tested em-
pirically. Kim and Cohen (2010) proved the relevance of physical and linguistic
distance by using panel methods in case of Western countries. Mayda (2010)
and Sprenger (2013) demonstrated the importance of distance, costs of border
crossing, language similarities and common cultural features while stressing the
facilitating effect of institutions that guarantee the conditions of free labor flow.
Palmer and Pytlikova (2015) used panel methods on European data and found a
causal relationship between labor market regulations of receiving countries and
the amount of immigration.
All of this can be formalized as follows. Consider the decision of a potential
migrant who can choose between two alternatives where the above-mentioned
costs and benefits are illustrated by proxy variables. Assume that their options
are to stay in the country of origin or to move abroad.3 For the sake of simplicity,
let us assume only one time period after migration where the related variables
represent discounted net present values of all future flows of costs and benefits.
As a result, in line with the aforementioned theoretical considerations and em-
pirical evidences, an agent compares equations (1) and (2) in their cost-benefit
calculation:
NBo  αo I o  Go  1  αo U o , (1)
NBr  αr I r  Gr  1  αr U r  Cm , (2)
where NB is net benefit of related alternatives, I is income obtained from work,
G is net public service level, U is unemployment benefit and Cm is mobility cost.
Furthermore, α is the probability of finding a job while o and r indices refer to
country of origin and receiving country, respectively. It implies that migration is
a rational choice for the agent if net benefit in the receiving country is higher than
that in the country of origin. That is:
NBr  NBo . (3)
For the sake of simplicity, assuming that actors certainly find a job in both
countries (that is αr = α0 = 1) and defining Δr = Ir – I0, ΔG = Gr – G0 gives the fol-
lowing expression:

ΔI + ΔG = Cm . (4)

3
Although it may be strange to assume that potential migrants have two alternatives, a rational
actor faces this kind of situation. Since a potential migrant with perfect information compares
their situation with the best foreign alternative, therefore it is not necessary to consider more
options. Nevertheless, it is a harsh constraint that may be loosened in further studies. For
application of Simon’s (1972) bounded rationality concept in migration theory, see Speare
(1974).

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

177 JÓZSEF GOLOVICS

This implies that international differences in wages and net public service
levels may be balanced by mobility costs. Although the operationalization and
measurement of the latter in monetary terms is not unambiguous, it is undeniable
that the amount of mobility costs dropped drastically in the last decades, thanks to
immense technological progress. It follows that nowadays’ extremely low mobil-
ity costs should incentivize rational actors towards migration in the presence of
huge economic disparities.
In this context, it should be stressed that the economic literature outlined above
is based on a mostly common idea: voluntary migration is driven by economic
motives and it can be treated as a migrant’s autonomous or individualistic deci-
sion (Collins 2018). And indeed, this approach undoubtedly explains a significant
part of international migration. However, new considerations emerged from other
disciplines of social sciences in order to understand contemporary migration in
greater depth (Carling – Collins 2018). Besides that, the empirical puzzle out-
lined in Section 1 remains unexplained: why do people stay in their countries of
origin despite the fact that migration would be beneficial for them in monetary
terms? We argue that the incorporation of emotional factors would help to explain
this ‘missing migration’. This is in line with Carling and Collin’s (2018: 913)
argument which claims that ‘drawing attention to emotions is critical to humanis-
ing migration’. Nevertheless, as opposed to their approach, we do this by extend-
ing the boundaries of the standard human capital model of migration. Existing
literature also foreshadows this direction. However, while several authors (e.g.
Bauer – Zimmermann 1995; DaVanzo 1981; De Jong – Fawcett, 1981; Massey et
al. 1993) acknowledge the importance of psychological costs (due to the separa-
tion from relatives) among non-monetary costs of migration, they do not go into
details about their ways of functionality. As an example, Sjaastad (1962) does
not treat them as real resource costs so he does not include them into the cost-
benefit calculation. He argues that psychological costs are hardy measurable and
are analogous to consumer surplus. Furthermore, it does not affect individual’s
choice above a certain level. Bodenhöfer (1967) uses the same argument as well.
However, we disagree with this argument. In Section 3, we build a model that
incorporates the actor’s psychological ties into the model and thus gives a real
sense for the psychological cost of migration.

3. THE MODEL

This model focuses on potential migrants’ migration decisions. In the paper, mi-
gration is defined as a voluntary cross-border movement with the aim of long-
term (more than one year) residential settlement. Our special interest concerns

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

 THE ROLE OF LOYALTY IN MIGRATION INTENTIONS 178

labor migration within the institutional framework of the European Union, which
is still international migration, even though the aim of the European integra-
tion is the creation of a single labor market. Anyway, from a theoretical point
of view, similar motivations lay behind both internal and international migra-
tion, and the only difference between them lies in their institutional framework
(Bauer – Zimmermann, 1995). While the EU has major achievements in the field
of the harmonization of formal institutions, legal differences do exist, and infor-
mal institutions, norms, cultural conventions are still very diverse. The defini-
tion above implies that commuting and other types of short-term movements are
omitted from this paper. Although these actions may also have similar driving
forces, psychological costs derived from emotional ties occur mostly in case of
long term migration.
In this context, our model is directly built on the one described in Section 2.
We thus follow the Sjaastadian tradition, which is called the neoclassical mi-
croeconomic theory of migration (Massey et al. 1993), standing on the basis of
methodological individualism. However, as it has been emphasized by other dis-
ciplines of social sciences, desire to migrate and the act of migration cannot be
understood without its wider social context (Collins 2018). Therefore, we put a
special emphasis on the social environment of the migration decision. Man is a
‘social creature’, thus people get in touch with each other, they influence each
other and they care about others. The circle of care might be narrowed in some
cases (e.g. restricted to the family), while others might care about their wider
environment (e.g. neighborhood or the whole society). It often results in people
not seeking their self-interest but considering the welfare of others, which is the
case of altruism.
Although altruism contradicts the basic behavioral assumption of economics,
it has become an ordinary concept – with a huge amount of empirical evidence
– in mainstream economics as well. It also means that several forms and inter-
pretations of it have been outlined in the economic literature (Golovics 2015). In
this study, altruism is introduced into the model by using Pollak’s (1976) inter-
dependent preferences concept. This manifestation of altruism was elaborated
in Becker’s (1974; 1976) work and also applied by Ishikawa (1975), Bolnick
(1979), Rapoport (1995) and Ottoni-Wilhelm et al. (2017). In these types of mod-
els, other people’s welfare may also be included in the altruist’s utility function,
which means that considering others’ interest is also part of the agent’s maximiza-
tion problem. However, it should be emphasized that interdependence of utility
functions is not derived from an externality generated by someone’s action but
based on interpersonal relations (Choi 1993; Hámori 2003).
Thus, in our model, a potential migrants’ welfare depends not only on their
own utility but also on changes in others’ utility related to them:

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

179 JÓZSEF GOLOVICS

U i  U i ( NBi ; γ x SU x ; γ y SU y ; γ z SU z ;), (5)

where Ui is the agent’s utiliy, NBi is their net benefit level resulting from the mi-
gration decision, ΔSU is the agent’s ‘social utility’ factor that embodies changes
in others’ utility about whom agent cares. Indices x, y, z refer to these related
people, while γ measures the strength of their relationship.
These types of altruistic relationships are defined as loyalty in this paper and it
can be applied towards any person or group. It signifies that we consider a person
loyal if they have any of this kind of altruistic relationship. Existence and the
extent of these altruistic relationships may vary among the members of society
but they are considered to be exogenously given for each individual at the time of
migration decision making.4
In this context, loyalty is a more complex factor than a simple type of mobility
cost, labelled as a psychological cost in the previous literature. The latter merely
refers to welfare loss generated by breaking-up interpersonal ties. However, the
loyalty concept of this study also takes the fact into account that loyal people
include each kind of change in related people’s welfare generated by their migra-
tion in their own utility function. The sign of change may also be positive in the
case when the aim of migration is to find a better job abroad in order to help fam-
ily left behind with remittances – which is also an altruistic and ‘loyal’ motive.
The loyalty concept defined above can be connected to Hirschman’s (1970)
‘exit-voice-loyalty theory’ as well. According to Hirschman, a person has two
choices in case of dissatisfaction (e.g. with an organization membership, con-
sumption of a product or living in a country): i) they can choose an exit strategy
by quitting the situation; or ii) they can give voice to their dissatisfaction while
staying in the situation. As Mike (2012: 59) stresses, ‘exit and voice may be
substitutes or complements’, hence dissatisfied individuals’ choice depends not
just on circumstances and characteristics of the situation but on their loyalty, too.
Thus, loyalty functions as an equilibrating factor inasmuch as its existence makes
the choice of exit less likely.
The applicability of Hirschman’s concept in migration theory has already been
raised in several papers, including Hirschman (1978) himself; however, none of
them reaches the stage of formal modelling. While Ruget and Usmanalieva (2008)
discuss the role of loyalty only in general terms, Schewel (2015) and Carling and
Schewel (2018) mention loyalty as a non-economic retaining factor, Newland

4
Although this paper treats loyalty as an exogenous factor for reasons of space, the existence
and the extent of loyalty could be connected to the individual’s previous life. For a detailed
discussion about the connection between loyalty and previous specific investments, see Mike
(2012).

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

 THE ROLE OF LOYALTY IN MIGRATION INTENTIONS 180

(2010), Johnston (2011) and Burgess (2012) examine the effect of emigrants on
the politics of their country of origin with qualitative tools, Hoffmann (2010)
inquires the empirical applicability and a possible extension of Hirschman’s ter-
minology in connection with international migration.
The identity of Hirschman’s loyalty concept with the one defined in this paper
is manifested in its effect on a migrant’s utility. In Hirschman’s (1970) theory,
loyal people do not ignore the fate of their former communities. And if they know
that their exit aggravates the community’s decline, they internalize it as an ‘exit
tax’.5 Effect of the latter is equivalent to that of loyalty defined in this paper: a
welfare change in the actor’s utility function. A related person’s psychological
pain generated by the cessation of daily connections or the negative fiscal ex-
ternality6 born by the society of the home country can be mentioned as types of
welfare losses, while the assisting force of remittances could be a form of wel-
fare gain in concern. Nonetheless, it should be emphasized that ‘loyalty tax’ in
our model is only triggered if related people’s welfare changes due to migration.
Thus, it does not come into effect if family members also migrate with the actor.
All of this can be incorporated into the model outlined above by introducing a
loyalty factor in a potential migrant’s utility function. Therefore, if the potential
migrant is loyal (i.e. at least one of γi ≠ 0 (where i = x, y, z, …) in expression (5)),
then loyalty factor L is added to their net benefits in case of migration which is
given (6) below.7 This loyalty factor L embodies the net of welfare changes from
all the altruistic relationships that the actor possesses.8

NBr  αr I r  Gr  (1  αr )U r  Cm  L (6)

where L   γi SU i and (i = x, y, z, …).

i

5
This concept is similar to Simon’s (1990) idea, who argued that society tends to tax docile
persons by making them perform altruistic acts.
6
Fiscal externality refers to a situation where local public goods are financed by local taxes.
In this case, residential decisions (i.e. moving to another municipality) change the tax base
in cities in concern and, as a result, local governments should change tax rates to maintain
the former level of public goods or equivalently, adjust the level of public goods provision to
the new level of tax revenues (Wildasin 1989; Hercowitz – Pines 1991). In case of a person’s
migration, this effect is marginal. However, in case of mass emigration, negative fiscal exter-
nality effects can be extremely dangerous.
7
It means that in the case of a disloyal person, L = 0.
8
Although, one might argue that the loyalty factor (multiplied either by the same or by another
parameter) should be included in the potential migrant’s net benefit function in case of staying
as well, this technique is not applied in this paper. It can be interpreted as follows: migration
may reduce individual’s welfare through their altruistic relations but staying does not provide
any welfare surplus.

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

181 JÓZSEF GOLOVICS

In this form, the loyalty factor represents a feedback from related persons’ wel-
fare changes through altruistic bonds. It can be interpreted in a way that a loyal
individual has such connections with other persons (e.g. friends, family members,
community, society) that changes in the latter’s welfare (either positive or nega-
tive), have a welfare effect on their own utility with the same sign. And although
loyalty is hard to operationalize or express in monetary terms, our proposition is
that loyalty can deter potential migrants from going abroad if:
I  G  Cm  L  0. (7)
This suggests that if the net of wage differences, public service level differenc-
es and mobility costs supplemented by welfare change expressed through loyalty
factor is negative, then potential migrants choose to stay in the country of origin.
Thus, loyalty can offset the attractiveness of migration in the presence of huge
development disparities and low mobility costs. As a result, it may explain the
“missing migration” puzzle raised in Section 1. Consequently, loyalty towards
the family, the community or the society of the country of origin may be the rea-
son why masses of EU citizens (mostly from new member states) do not choose
to live in another member state where the standard of living and labor market
opportunities are better, despite low mobility costs and institutionally guaranteed
freedom.

4. DATA AND METHODOLOGY

In order to empirically test our hypothesis about the role of loyalty in migration
decisions, we use Eurobarometer survey data and estimate a logistic regression
model. The main characteristics of the dataset and the exact specification of em-
pirical model are detailed below.
In compliance with our microeconomic theoretic model, micro level data is
used in empirical tests in order to avoid ecological fallacy. We use individual
data from Eurobarometer 73.3 (Mar-Apr 2010), which is an EU-wide representa-
tive survey that contains answers from 26,602 respondents (European Commis-
sion 2012). The survey comprises ordinary questions about respondents’ socio-
economic backgrounds, while it contains thematic blocks as well. One of these
thematic blocks was about the respondent’s mobility history and intentions. This
wave of Eurobarometer surveys included questions about the respondents’ at-
tachment to different entities as well. This combination of questions makes it
possible to empirically test our migration model supplemented with loyalty while
controlling for standard socio-economic variables.

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

 THE ROLE OF LOYALTY IN MIGRATION INTENTIONS 182

Our dependent variable is based on the respondents’ answer about their mi-
gration intention in the next ten years.9 Although statements about future inten-
tions are not equivalent to actual migration, using migration intention data in
empirical studies is an ordinary practice in migration research (see e.g. Fouarge
– Ester 2008; Sik – Szeitl 2016) because of the “high correlation between mi-
gration intentions and their realization” (Zaiceva – Zimmermann 2008a: 439).
Blanchflower et al. (2007) and Zaiceva and Zimmermann (2008b) showed that
characteristics of people who intend to migrate and those of actual migrants are
quite similar. Sik and Szeitl (2016) confirmed these similarities. Furthermore,
Böheim and Taylor (2002) found that people with previous preference to move
are far more likely to go abroad than others. There are practical reasons too for
using migration intention data. First, researchers face serious difficulties in con-
ceptualization, measurement and registration of migration (for details see Bauer
– Zimmermann 1995), which results in a lack of cross-country micro data about
migration, therefore finding a reliable and representative dataset that includes
personal attributes of migrants and contains these pieces of information about a
representative reference group, too, (i.e. people who did not emigrate) is hardly
possible. Secondly, as Bauer and Zimmermann (1995) emphasize, actual migrants
represent only a very small part of society which may raise problems in estima-
tion, therefore using migration intention data may lighten this inconvenience (i.e.
people with migration intentions represent a bigger fraction of society). Conse-
quently, although migration intention data overestimates potential migration, it
can be considered as a sufficient tool to empirically analyze motivations behind
migration decisions.
In our empirical specification the dependent variable is a dummy, which means
that we grouped respondents’ answer as follows: 1 if they intend to migrate (very
likely, fairly likely), 0 if they do not intend to migrate (not very likely, not at
all likely). If the dependent variable equals 1, then the person is defined to be
a potential migrant. Figure 1 reports the proportion of potential migrants per
country.
Our focal independent variables serve as proxies for the different aspects of a
person’s loyalty. For that, we use answers for survey questions about the respond-
ents’ attachment to their countries (Loyalty/country) and to their cities/towns/
villages (Loyalty/city). Answers are coded as follows: 1: not attached at all; 2: not
very attached; 3: fairly attached; 4: very attached. We expect that stronger attach-
ment represents stronger altruism towards the community or society of the entity
in concern, which means that one has less intention to migrate.

9
The exact wording of the question is the following: ‘How likely do you think it is that you will
move to another country within the next ten years, to live there?’

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

183 JÓZSEF GOLOVICS

Figure 1. Proportion of potential migrants in EU member states (2010)

Source: European Commission (2012).

Dummies indicating marital status are also used as independent variables (Liv-
ing with partner, Single, Divorced, Widow, where reference group is Married).
Number of children under 15 (Children) is also an important characteristic. Both
Becker (1974) and Simon (1993) claim that family is the basis of altruistic rela-
tionships because of sociobiological reasons, so we expect that people with big-
ger and institutionalized families are less likely to migrate.
A respondent’s minority identity in religious, cultural or ethnic dimension
(Minority) is another independent dummy variable. Simon (1993) claims that mi-
nority groups tend to teach their docile members to behave altruistically towards
other group members, thus, the feeling of belonging to a minority group in terms
of religion, culture or ethnic origin is expected to mitigate migration intentions.
Age is a common control variable in a microeconomic migration model since
older people have a shorter time horizon on which they can maximize the lifetime
welfare gain of migration. Besides that, Lowe and Ritchey’s (1973) empirical
investigation found that older people have more altruistic attitudes than young-
er people, consequently, there may be a correlation between age and altruistic

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

 THE ROLE OF LOYALTY IN MIGRATION INTENTIONS 184

loyalty, so it can be assumed that an increase in age decreases the probability of

migration.
Furthermore, we use further control variables that are common in migration
literature (see Fouarge – Ester 2008; Sik – Szeitl 2016 and Zaiceva – Zimmer-
mann 2008a). Homeowners and homeowners with mortgage are expected to be
less likely to migrate since their houses serve as a location specific investment.
More-educated people (Education) are expected to have higher migration inten-
tions as their human capital is utilized easier in foreign countries too, so they
have a higher probability of finding a job. The same argument can be applied
for those people who are fluent in a foreign language (Language). Foreign lan-
guage speakers can also seek information with less transaction costs. Internet
usage (Internet) also eases information gathering so people who use the internet
more often are expected to have higher migratory potentials. Since migration
networks facilitate migration significantly (Massey 1990), it can be expected
that people with relatives (Network/relatives) or friends (Network/friends)
abroad are more likely to migrate. The same is expected for those who have
previous migratory history: i.e. worked abroad (Network/work), studied abroad
(Network/study), lived abroad (Network/live) or regularly go abroad (Network/
go). To control financial status, a dummy variable (Financial status) is used
that indicates whether the respondent had difficulties in paying their bills at the
end of the month in the last twelve months. Because of sociological reasons, we
also control for respondents’ gender (Male), inhabitancy (Small town and Large
town) and labor market status (Unemployed, Retired, Student, Self-employed).
Finally, a dummy (EU15) is used to control for different macroeconomic and
institutional backgrounds of old and new member states. Descriptive statistics
for the variables are in Appendix Table A1.
In order to estimate the model, a logistic regression is used. Since our depend-
ent variable is qualitative, simple OLS assumptions do not hold. Because of this,
the usage of OLS (a so called linear probability model) may predict probabilities
higher than 1. However, based on its properties, a logit model is an appropri-
ate one for our estimation. Because of the problem of heteroscedasticity, robust
standard errors are used. Our empirical strategy is to estimate a basic model
(Specification 1) with ordinary independent variables and another one with sup-
plemented proxies of loyalty (Specification 2). Eventually, we compare them.
Goodness of fit is measured by the models’ pseudo R-squared and their ability to
correctly predict values.

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

185 JÓZSEF GOLOVICS

5. RESULTS

Table 1 presents results from estimating Specification 1 and Specification 2. It

shows marginal effects for the migration intentions related to the intentions to
stay in the respondents’ home country.
Results show that most of our variables turned to be significant in both speci-
fications and they are in line with general human capital theory of migration.
Signs and magnitudes of coefficients are similar to those of previous studies (e.g.
Fouarge – Ester 2008; Sik – Szeitl 2016; Zaiceva – Zimmermann 2008a) as well.
Older people are less likely to migrate and owning a house (either with mortgage
or without it) also decreases the willingness to migrate. Potential migrants are
positively selected with respect to education, language knowledge and internet
usage. Network effects also matter: having relatives or friends abroad and previ-
ous foreign experiences increase the probability of migration. Those who have
difficulties with paying the bills are more willing to migrate, which indicates
that financial reasons may be important driving forces of migration. Willingness
to migrate is higher among male respondents and people living in towns. Labor
market status variables are also in line with expectations: unemployed people and
students are more likely to migrate, while retired respondents have lower migra-
tion intentions. Being employed does not have a significant effect on the willing-
ness to migrate compared to being self-employed in either specification. Citizens
from old EU member states are less willing to migrate, which can be explained
by better macroeconomic conditions, standards of living as well as different in-
stitutional backgrounds.
Besides that, loyalty also matters in shaping migration intentions. People who
are loyal towards their country or city are significantly less willing to migrate.
In this regard, loyalty towards country has a higher impact. Altruistic considera-
tions towards family members may also matter. People with more children are
less likely to migrate, while being single or divorced increases the probability
of migration. Living with a partner or being widowed has a statistically insig-
nificant effect on migration intentions compared to that of married respondents.
Belonging to a minority group turned out to be significant, however, its impact on
willingness to migrate contradicts our expectations: members of minority groups
have larger intentions to migrate. The reason behind it may be complex. On the
one hand, as DaVanzo (1981) emphasizes, former migration experiences foster
repeated migration. Thus, the supposition that members of minority groups may
be immigrants can be an explanation for their higher willingness to migrate. The
fact that minorities tend to be displaced to the periphery of society may also
enhance migration intentions. In this respect, note that altruism may foster emi-

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

 THE ROLE OF LOYALTY IN MIGRATION INTENTIONS 186

Table 1. Determinants of migration intentions in the EU. Marginal effects from Logit model
Specification 1 Specification 2
Migration intention
Loyalty/country –0.0252*** (0.00238)
Loyalty/city –0.0118*** (0.00206)
Living with partner 0.00850 (0.00518)
Single 0.0222*** (0.00500)
Divorced 0.0168** (0.00735)
Widow 0.00367 (0.0107)
Children –0.00959*** (0.00179)
Minority 0.0278*** (0.00566)
Age –0.00298*** (0.000145) –0.00260*** (0.000165)
Homeowner –0.0234*** (0.00366) –0.0134*** (0.00384)
Homeowner with
mortgage –0.0253*** (0.00313) –0.0166*** (0.00349)
Education 0.000826** (0.000321) 0.000709** (0.000327)
Language 0.0180*** (0.00319) 0.0121*** (0.00332)
Internet 0.00309*** (0.000736) 0.00365*** (0.000809)
Network/relatives 0.0239*** (0.00341) 0.0235*** (0.00363)
Network/friends 0.0524*** (0.00349) 0.0499*** (0.00371)
Network/work 0.0434*** (0.00555) 0.0421*** (0.00584)
Network/study 0.0187*** (0.00580) 0.0120** (0.00579)
Network/live 0.0408*** (0.00648) 0.0338*** (0.00661)
Network/go 0.0180*** (0.00383) 0.0119*** (0.00384)
Financial status 0.0271*** (0.00326) 0.0255*** (0.00353)
Male 0.0184*** (0.00290) 0.0125*** (0.00306)
Small town 0.0120*** (0.00358) 0.00856** (0.00374)
Large town 0.0171*** (0.00387) 0.0118*** (0.00403)
Unemployed 0.0201*** (0.00701) 0.0164** (0.00734)
Retired –0.0171** (0.00688) –0.0210*** (0.00702)
Student 0.0745*** (0.0117) 0.0483*** (0.0109)
Employed 0.00164 (0.00536) –0.00226 (0.00554)
EU15 –0.00958*** (0.00322) –0.0114*** (0.00350)

Pseudo R2 0.2385 0.2649

Observations 25,230 20,139

Notes: Robust standard errors in parentheses. Significances: *** p<0.01, ** p<0.05, * p<0.1. Reference groups
are married, not owning a house, not feeling belonging to a minority group, not speaking any language flu-
ently, not having any network abroad, not having financial difficulties at paying the bills, female, rural area,
self-employed.
Source: author’s calculations based on European Commission (2012).

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

187 JÓZSEF GOLOVICS

gration if the individual would be able to help the family by moving to another
country which is in line with our theoretical model.
As a result, loyalty contributes to increasing the explanatory power of the
standard microeconomic model of migration. By adding proxies of loyalty, no
standard variable changes its sign or loses its significance. Pseudo R-squared of
Specification 2 is 2.64 percentage points higher than that of Specification 1, which
is a moderate difference, however, note that Logit models do not tend to have
high Pseudo R-squared values in empirical studies.10 Concerning the predictive
power of the model, Specification 2 is able to correctly classify 88.27 percentage
of observations by using a cutoff point at 0.5 (which is also a moderate increase
if we consider that Specification 1 is able to correctly classify 87.44 percentage
of observations). Nevertheless, incorporating loyalty into the model enhances its
explanatory power even if it is a slight increase.
Although these results support the argument that loyalty matters in shaping
migration intentions, our empirical investigation has limitations. Survey data has
been taken in 2010, after the financial crisis, so special distortion may be present
in the data set. Besides that, although using migration intention data is an ordi-
nary practice in migration literature, it has to be emphasized that articulation
of intention to migrate is only a necessary but not sufficient condition to actual
migration. As Carling and Schewel (2018) and Gödri and Feleky (2017) argue,
certain barriers may prevent actual migration even if one aspires to move. Conse-
quently, further research – if it is possible – is needed on actual migration data in
order to test the robustness of our results.

6. CONCLUSIONS

This paper focused on the determinants of individual migration decisions. Al-

though neoclassical human capital theory of migration has major achievements
in this field, its explanatory power is unsatisfactory since it ignores many psycho-
logical aspects of the decision as well as the well-known fact that people feel like
belonging to a certain community. Thus, this paper aimed at building a theoretical
framework that incorporated loyalty into the microeconomic theory of migration.
In this regard, a plausible and well-grounded theoretical model was built that was
able to provide a reason for the unexplained ‘missing’ migration in the European
Union.

10
See Sik and Szeitl (2016) or Zaiceva and Zimmermann (2008a), where pseudo R-squared
values tend to move between 0.10 and 0.21.

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

 THE ROLE OF LOYALTY IN MIGRATION INTENTIONS 188

Our model was tested empirically on migration intentions data. Although vari-
ables used in the empirical model are not entirely novel in migration studies, this
time they were supported by a solid economic model that is perfectly compat-
ible with neoclassical human capital theory. Estimations show that loyalty indeed
matters in shaping people’s willingness to migrate, however, robustness of results
could be enhanced if better data were available. Nonetheless, the analysis in this
paper showed that the presence of altruistic ties and loyalty towards individuals’
community may make them to stay in their country of origin even though migra-
tion would be a rational alternative from a financial point of view.
Results of this paper open new perspectives in migration studies. On the one
hand, our hypothesis could be tested on other types of datasets (e.g. about actual
migration or with higher variance in the dependent variable). On the other hand,
it sheds another light on existing literature. In line with migration network theory
(Massey 1990), altruistic relationships defined in this paper may also strengthen
one’s willingness to migrate if one’s loved ones have already migrated. However,
it seems to contradict the theory outlined above, and the truth is that ‘net util-
ity’ from emotional bonds play an important role in migration choices. Finally,
this paper re-opens the question about self-interest and raises questions about
the origins of altruism in order to understand the very nature of human decision-
making.

REFERENCES

Ahmad, N. – Hussain, Z. – Sial, M. H. – Hussain, I. – Akram, W. (2008): Macroeconomic Deter-

minants of International Migration from Pakistan. Pakistan Economic and Social Review 46(2):
85–99.
Andor, L. (2014): Labour Mobility in the European Union – The Inconvenient Truth. Lecture at
University of Bristol, Bristol. http://europa.eu/rapid/press-release_SPEECH-14-115_en.htm,
accessed 13/11/2017.
Bauer, T. – Zimmermann, K. (1995): Modelling International Migration: Economic and Economet-
ric Issues. In: van der Erf, R. – Heering, L. (eds): Causes of International Migration. Proceed-
ings of a Workshop. Luxembourg: Eurostat, pp. 95–115.
Becker, G. S. (1962): Investment in Human Capital: A Theoretical Analysis. Journal of Political
Economy 70(5): 9–49.
Becker, G. S. (1974): A Theory of Social Interactions. Journal of Political Economy 82(6): 1063–
1093.
Becker, G. S. (1976): Altruism, Egoism, and Genetic Fitness: Economics and Sociobiology. Jour-
nal of Economic Literature 14(3): 817–826.
Blanchflower, D. G.– Saleheen, J. – Shadforth, C. (2007): The Impact of the Recent Migration from
Eastern Europe on the UK Economy. IZA Discussion Paper No. 2615.
Bodenhöfer, H. J. (1967): The Mobility of Labor and the Theory of Human Capital. The Journal of
Human Resources 2(4): 431–448.

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

189 JÓZSEF GOLOVICS

Bolnick, B. R. (1979): Government as a Super Becker-Altruist. Public Choice 34(3): 499–504.

Böheim, R. – Taylor, M. P. (2002): Tied Down or Room to Move? Investigating the Relation-
ships between Housing Tenure, Employment Status and Residential Mobility in Britain. Scottish
Journal of Political Economy 49(4): 369–392.
Burgess, K. (2012): Migrants, Remittances, and Politics: Loyalty and Voice after Exit. The Fletcher
Forum of World Affairs 36(1): 43–55.
Carling, J. – Collins, F. (2018): Aspiration, Desire and Drivers of Migration. Journal of Ethnic and
Migration Studies 44(6): 909–926.
Carling, J. – Schewel, K. (2018): Revisiting Aspiration and Ability in International Migration. Jour-
nal of Ethnic and Migration Studies 44(6): 945–963.
Choi, Y. B. (1993): Paradigms and Conventions. Uncertainty, Decision Making, and Entrepreneur-
ship. Ann Arbor: The Universitiy of Michigan Press.
Collins, F. L. (2018): Desire as a Theory for Migration Studies: Temporality, Assemblage and Be-
coming in the Narratives of Migrants. Journal of Ethnic and Migration Studies 44(6): 964–
980.
DaVanzo, J. (1981): Microeconomic Approaches to Studying Migration Decisions. In: De Jong, G.
F. – Gardner, R. W. (eds): Migration Decision Making. Multidisciplinary Approaches to Mic-
rolevel Studies in Developed and Developing Countries. New York: Pergamon Press, pp. 90–
129.
De Jong, G. F. – Fawcett, J. T. (1981): Motivations for Migration: An Assessment and a Value-
Expectancy Research Model. In: De Jong, G. F. – Gardner, R. W. (eds): Migration Decision
Making. Multidisciplinary Approaches to Microlevel Studies in Developed and Developing
Countries. New York: Pergamon Press, pp. 13–58.
European Commission (2012): Eurobarometer 73.3 (Mar-Apr 2010). TNS OPINION & SOCIAL,
Brussels [Producer]. GESIS Data Archive, Cologne. ZA5233 Data file Version 3.0.0. https://dbk.
gesis.org/dbksearch/sdesc2.asp?no=5233&db=e&doi=10.4232/1.10114, accessed 01/12/2016.
Eurostat (2017): Annual Net Earnings (earn_nt_net). Eurostat Database. http://appsso.eurostat.
ec.europa.eu/nui/show.do?dataset=earn_nt_net, accessed 02/12/2017.
Eurostat (2018): Population on 1 January by Five Year Age Group, Sex and Citizenship [migr_
pop1ctz]. Eurostat Database. http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=migr_
pop1ctz&lang=en, accessed 09/02/2018.
Fouarge, D. – Ester, P. (2008): How Willing are Europeans to Migrate? A Comparison of Migration
Intentions in Western and Eastern Europe. In: Ester, P. – Muffels, R. – Schippers, J. – Wilthagen,
T. (eds): Innovating European Labour Markets. Cheltenham: Edward Elgar, pp. 49–71.
Geis, W. – Uebelmesser, S. – Werding, M. (2008): How do Migrants Choose Their Destination
Country? CESifo Working Paper No. 2506.
Gödri, I. – Feleky, A. (2017): Selection of Migrants and Relization of Migration Intentions – Les-
sons from a Panel Study. Hungarian Demographic Research Institute Working Papers on Popu-
lation, Family and Welfare No. 29.
Golovics, J. (2015): Bounded Rationality and Altruism: Behaviourism in Economics. Financial
and Economic Review 14(2): 158–172.
Hámori, B. (2003): Érzelemgazdaságtan [Economics of Emotions]. Budapest: Kossuth Kiadó.
Hart, R. A. (1975): Interregional Economic Migration: Some Theoretical Considerations (Part II).
Journal of Regional Science 15(3): 289–300.
Hercowitz, Z. – Pines, D. (1991): Migration with Fiscal Externalities. Journal of Public Economics
46(2): 163–180.
Hirschman, A. O. (1970): Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations,
and States. Cambridge: Harvard University Press.

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

 THE ROLE OF LOYALTY IN MIGRATION INTENTIONS 190

Hirschman, A. O. (1978): Exit, Voice, and the State. World Politics 31(1): 90–107.
Hoffmann, B. (2010): Bringing Hirschman Back In: “Exit”, “Voice”, and “Loyalty” in the Politics
of Transnational Migration. The Latin Americanist 54(2): 57–73.
Ishikawa, T. (1975): Family Structures and Family Values in the Theory of Income Distribution.
Journal of Political Economy 83(5): 987–1008.
Johnston, T. (2011): Voice after Exit? Remittances and Migration across Authoritarian Regimes.
UM Comparative Politics Workshop, January 28, 2011. https://sites.lsa.umich.edu/tsebelis/wp-
content/uploads/sites/223/2015/01/johnston.cpw2011.pdf, accessed 02/11/2017.
Kim, K. – Cohen, J. E. (2010): Determinants of International Migration Flows to and from Indus-
trialized Countries: A Panel Data Approach Beyond Gravity. International Migration Review
44(4): 899–932.
Kureková, L. (2013): Welfare Systems as Emigration Factor: Evidence from the New Accession
States. Journal of Common Market Studies 51(4): 721–739.
Lowe, R. – Ritchey, G. (1973): Relation of Altruism to Age, Social Class, and Ethnic Identity. Psy-
chological Reports 33(2): 567–572.
Massey, D. S. (1990): Social Structure, Household Strategies, and the Cumulative Causation of
Migration. Population Index 56(1): 3–26.
Massey, D. S. – Arango, J. – Hugo, G. – Kouaouci, A. – Pellegrino, A. – Taylor, E. J. (1993): Theo-
ries of International Migration: A Review and Appraisal. Population and Development Review
19(3): 431–466.
Mayda, A. M. (2010): International Migration: A Panel Data Analysis of the Determinants of Bilat-
eral Flows. Journal of Population Economics 23(4): 1249–1274.
Mike, K. (2012): Hirschman Meets Williamson: Relationship-Specific Investment and Loyalty.
Corvinus Journal of Sociology and Social Policy 3(2): 55–81.
Newland, K. (2010): Voice After Exit: Diaspora Advocacy. Washington, DC: Migration Policy In-
stitute. https://www.migrationpolicy.org/pubs/diasporas-advocacy.pdf, accessed 02/11/2017.
Ottoni-Wilhelm, M. – Vesterlund, L. – Xie, H. (2017): Why do People Give? Testing Pure and
Impure Altruism. American Economic Review 107(11): 3617–3633.
Palmer, J. – Pytlikova, M. (2015): Labor Market Laws and Intra-European Migration: The Role of
the State in Shaping Destination Choices. European Journal of Population 31(2): 127–153.
Pollak, R. A. (1976): Interdependent Preferences. The American Economic Review 66(3): 309–
320.
Rapoport, H. (1995): Coordination, Altruism and Under-Development. Kyklos 48(3): 389–407.
Ruget, V. – Usmanalieva, B. (2008): Citizenship, Migration and Loyalty towards the State: A Case
Study of the Kyrgyzstani Migrants Working in Russia and Kazakhstan. Central Asian Survey
2: 129–141.
Schewel, K. (2015): Understanding the Aspiration to Stay: A Case Study of Young Adults in
Senegal. International Migration Institute Working Papers 107.
Sik, E. – Szeitl, B. (2016): Migration Intentions in Contemporary Hungary. In: Blaskó, Z. – Faze-
kas, K. (eds): The Hungarian Labour Market 2016. Budapest: Institute of Economics, Centre
for Economic and Regional Studies, Hungarian Academy of Science, pp. 55–59.
Simon, H. A. (1972): Theories of Bounded Rationality. In: McGuire, C. B. – Radner, R. (eds):
Decision and Organization. Amsterdam: North-Holland Publishing Company, pp. 161–176.
Simon, H. A. (1990): A Mechanism for Social Selection and Successful Altruism. Science 250:
1665–1668.
Simon, H. A. (1993): Altruism and Economics. AEA Papers and Proceedings 82(2): 156–172.
Sjaastad, L. A. (1962): The Costs and Returns of Human Migration. Journal of Political Economy
70(5): 80–93.

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

191 JÓZSEF GOLOVICS

Speare, A. J. (1974): Residential Satisfaction as an Intervening Variable in Residential Mobility.

Demography 11(2): 173–188.
Sprenger, E. (2013): The Determinants of International Migration in the European Union: An Em-
pirical Analysis. IOS Working Papers 325.
Todaro, M. P. – Maruszko, L. (1987): Illegal Migration and US Immigration Reform: A Conceptual
Framework. Population and Development Review 13(1): 101–114.
UN (2017): International Migration Report 2017. New York: United Nations.
Wildasin, D. E. (1989): Interjurisdictional Capital Mobility: Fiscal Externality and a Corrective
Subsidy. Journal of Urban Economics 25: 193–212.
Zaiceva, A. – Zimmermann, K. F. (2008a): Scale, Diversity, and Determinants of Labour Migration
in Europe. Oxford Review of Economic Policy 24(3): 428–452.
Zaiceva, A., – Zimmermann, K. F. (2008b): Scale, Diversity, and Determinants of Labour Migra-
tion in Europe. IZA Discussion Paper No. 3595.

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC

 THE ROLE OF LOYALTY IN MIGRATION INTENTIONS 192

APPENDIX

Table A1. Descriptive statistics

Variable Observations Mean Std. Dev. Min Max
Dependent variable
Migration intention 26602 .1363807 .3431988 0 1

Independent variables
Loyalty\country 26520 3.54868 .630651 1 4
Loyalty\city 26527 3.376371 .7580354 1 4
Living with partner 26571 .0925069 .2897456 0 1
Single 26571 .1943096 .3956757 0 1
Divorced 26571 .0745173 .2626159 0 1
Widow 26571 .0905498 .2869733 0 1
Children 26602 .4465078 .8812927 0 14
Minority 21132 .0998012 .2997419 0 1
Age 26602 47.97042 18.20096 15 97
Homeowner 26602 .5281182 .4992181 0 1
Homeowner with mortgage 26602 .231975 .4221011 0 1
Education 26088 18.81923 5.087374 0 90
Language 26518 .4530508 .4978003 0 1
Internet 26602 3.523682 2.530618 0 6
Network\relatives 26525 .3295005 .4700407 0 1
Network\friends 26522 .4529824 .4977938 0 1
Network\work 26552 .1402531 .347256 0 1
Network\study 26557 .0809203 .2727178 0 1
Network\live 26556 .1056258 .3073639 0 1
Network\go 26523 .2126456 .4091867 0 1
Financial status 25974 .3975129 .4893931 0 1
Male 26602 .4608676 .4984757 0 1
Small town 26582 .3544128 .478344 0 1
Large town 26582 .2818449 .4499066 0 1
Unemployed 26602 .1641982 .3704623 0 1
Retired 26602 .2877227 .4527097 0 1
Student 26602 .082287 .2748067 0 1
Employed 26602 .3943313 .4887158 0 1
Source: author’s calculations based on European Commission (2012).

Open Access. This is an open-access article distributed under the terms of the Creative
Commons Attribution 4.0 International License (https://creativecommons.org/licenses/
by/4.0), which permits unrestricted use, distribution, and reproduction in any medium,
provided the original author and source are credited, a link to the CC License is provided,
and changes – if any – are indicated. (SID_1)

Society and Economy 42 (2020)

Unauthenticated | Downloaded 01/09/25 11:21 PM UTC