Introduction
As a result of globalisation and increasing immigration, there is increased interest in the economic activity of migrants (Ram and Jones 2008; Ivlevs and King 2012; Beckers and Blumberg, 2013; Elo 2016). Indeed, the rapid increase of international immigration in developed countries and their contribution to economic growth and regional development has encouraged studies on migrants and their participation in labour markets, not only as employees but also as entrepreneurs (Aliaga-Isla and Rialp 2013). Migrant entrepreneurs are able to bridge international boundaries while facing numerous complexities including different business contexts, borders and transnational identities and characteristics (Brinkerhoff 2016), and at the same time are able to utilise their entrepreneurial resources gained through internationalisation processes (Brinkerhoff 2009; Terjesen and Elam 2009). As part of their resource endowment, entrepreneurs can utilise human capital, which refers to the characteristics, skills, competences and education that are individual endowments, as well as social capital, which refers to the nature of social relations and how they can be drawn upon for individual benefit (Williams et al. 2017). Human capital and access to networks which generate social capital are important for indigenous entrepreneurs as well as migrants (Bagwell 2015).
Yet, despite its emergence as a research field (Brinkerhoff 2009; Çavusgil et al. 2011; Riddle and Brinkerhoff 2011; Elo et al. 2015), migrant entrepreneurship, and the resources they utilise, is still relatively under-researched (Peroni et al. 2016). There is some existing research which examines forced migration in the context of refugees and entrepreneurship (see, for example, Wauters and Lambrecht 2006, 2008; Lyon et al. 2008), which shows how different forms of capital can provide migrant entrepreneurs with access to a flow of resources, new market opportunities and business ideas (Bagwell 2015). However, there remains a need for more research on evaluating the ways in which migrant entrepreneurs contribute to the economy and patterns of diaspora business ownership (Fairlie and Lofstrom 2014). As a result, our paper examines the entrepreneurial activity of migrant entrepreneurs, with a particular focus on migrants who have been forced to migrate due to conflict. Economies experiencing conflict experience significant levels of forced migration as a result of war as well as ongoing economic and demographic challenges following the cessation of war. There is an established literature on economic migrants and their impacts on the country of residence (COR) (Ram et al. 2008; Vershinina et al. 2011; Barrett and Vershinina 2016); yet, there is a paucity of literature which examines those who are forced to migrate, and the role that human and social capital resources may play in them undertaking entrepreneurial activity.
Anzeige
As such, we fill a gap in the literature through a focus on identifying the factors which drive entrepreneurship among forced migrants from conflict-affected economies. In this study, we define entrepreneurs as those involved in the act of establishing a business or who currently own/manage a business (Williams and Williams 2012). The key research question is ‘How do human and social capital resources influence the entrepreneurial activity of migrants, with special reference to migrants who have been forced to migrate?’ Our contribution is to show how forced migrants harness their human and social capital resources in order to launch entrepreneurial ventures. We show that forced migration can have a direct impact on the resource endowments of forced migrants, limiting their short-term resources. However, over time, the probability of being an entrepreneur increases as new resources are added, including fluency of the language of the host country. Yet, despite the importance attached to social capital in the entrepreneurship, we find that connection to their ethnic group (denoting co-ethnic networks) has no effect on the probability of being an entrepreneur.
Our paper focuses on the post-conflict context of Kosovo, which has seen significant emigration to other parts of the world, particularly as a result of the wars of the 1990s (Adamson 2005), the break-up of the former Yugoslavia and subsequent declaration of independence in 2008, as well as ongoing economic and social challenges (Ivlevs and King 2012; Möllers et al. 2015; Vorley and Williams 2017). This emigration has left Kosovo with a sizable diaspora community spread around the world. The study uses Riinvest Migrant’s Survey data collected at the end of 2008 and beginning of 2009. The data utilises interviews with migrants who were returning to Kosovo immediately following the declaration of independence to examine the types of activities they were undertaking in their COR. As such, they maintain international links with their homeland, although they are not necessarily diaspora entrepreneurs, defined as those who establish entrepreneurial activities that span the national business environments of their home and host countries (Riddle et al. 2010), as all or the majority of their entrepreneurial activity takes place in their COR. The analysis shows that networking among the migrant entrepreneurs is a critical factor impacting entrepreneurial activity, although networking among their own ethnic community is not significant. Length of residency in the COR increases the probability of entrepreneurship, while educational attainment levels do not have an influence on entrepreneurship, although specific business training is important.
The remainder of the paper is structured as follows. First, the literature on the role and impact of migrant entrepreneurs is set out, followed by an examination of forced migration. Next, we set out the context of Kosovo, before the analysis of the survey is presented. The paper concludes by making a number of contributions to scholarship on diaspora entrepreneurship and forced migration.
Literature review
The role of migrant entrepreneurs
Entrepreneurship and migration are an increasing focus of academic literature and policy discourse. With over 232 million cross-border migrants, the development impacts of both migrants and entrepreneurs are likely to be substantial (Naude et al. 2015). Many migrants engage in entrepreneurial activities; however, these individuals have been largely peripheral in mainstream discourses on entrepreneurship (Ram et al. 2008). An emerging body of literature has begun to explore the experiences of such entrepreneurs (Anderson et al. 2010; Chand and Ghorbani 2011; Koning and Verver 2013), and this paper contributes to this discourse, through a focus on forced migrant entrepreneurship.
Anzeige
A key element of migrant entrepreneurship research has been to show a higher than average rate of entrepreneurial activity, a recurring feature applying just as much to the recent waves of Somalis and Poles as to their South Asian and Hong Kong Chinese precursors (Edwards et al. 2016). Neville et al. (2014) find that migrant enterprises are not generally better performing than those of non-migrants and that very often immigrant-owned firms underperformed; migrant firms who export are found to perform better. Within this, typical measures in firm performance such as experience, skills, gender, access to finance and growth orientation were found to apply in equal measure to both migrant and non-migrant enterprises (Naude et al. 2015). Yet, such groups have rarely figured in contemporary debates on entrepreneurship, other than in a few notable studies (Edwards et al. 2016; Ram et al. 2008). Indeed, given the diversity of ethnic and migrant entrepreneurship, it has been argued that research and practice have entered a phase of ‘super-diversity’, which aims to underling of complexity surpassing what has previously been experienced, and is distinguished by a dynamic interplay of variables among an increased number of new, small and scattered, multiple-origin, transnationally connected, socio-economically differentiated and legally stratified immigrants (Vertovec 2007).
As Ram et al. (2017) state, this new ‘super-diversity’ has presented challenges for researchers, for example, the unprecedented geographical hyper-mobility of the post-Soviet East Europeans (Legrain 2007) and the active transnational networks of African groups (Thompson 2016). At the same time, however, such changes do not generally translate into changes in performance, with new migrant business owners experiencing structural disadvantages similar to their predecessors (Edwards et al. 2016; Jones et al. 2014).
In common with entrepreneurs more broadly who can enter activity out of opportunity or necessity (Williams and Williams 2014), migrants arriving in a new country can be both pushed and pulled into entrepreneurship. With regards to push dynamics, migrants can face a range of labour market obstacles, which leaves entrepreneurial activity as their only option for generating income (Gilad and Levine 1986; Elo 2016). Such individuals can face discrimination from the indigenous population, lack relevant forms of cultural capital (knowledge and language skills in the host country) and also to access to host-country relevant social capital (Neville et al. 2014). For these individuals, developing forms of entrepreneurial activities is viewed as a survival strategy (Portes 1994), which involves relying on the social capital of their ethnic group (Drori et al. 2009) in the absence of any other relevant economic options. With regards to push dynamics, migrant groups can also be characterised as having a strong sense of self-sufficiency (Legrain 2007), which pulls them into forms of entrepreneurship as a means of ‘getting by and getting on’ (Anderson et al. 2010). Indeed, those with highly developed skills and numerous alternatives for their career and livelihood who embark on entrepreneurial endeavours (Mullings 2011), as well as those who are not pushed to migrate by necessity, but who choose to become migrants to fulfill their entrepreneurial, career or social goals and dreams (Elo 2016). Yet, the push/pull dichotomy can be considered to be simplistic, as often motivations will shift over time and will be dependent on context (Williams and Williams 2012). Indeed, in the case of forced migrants as the resources that they can access change over time, we posit that this will influence their motivations. While forced migrants may enter entrepreneurship out of necessity initially, as they acquire more resources, in particular through the development of human and social capital, they will be exposed to more opportunity-driven entrepreneurship.
Whether they enter entrepreneurship can also be determined by ethnicity (Fairlie and Meyer 1996). Ethnic characteristics can act as drivers for some migrant groups to be more entrepreneurial than others, including the ability to develop entrepreneurial strategies effectively (Chaganti and Greene 2002), or use their ethnic resources (i.e. access to social capital within an embedded network) in the new country (Chand and Ghorbani 2011; Koning and Verver 2013). As such, migrants can exploit the underdeveloped markets which have emerged within ‘ethnic enclaves’ in the host country (Koning and Verver 2013), using their access to ethnic and migrant networks and access to informal sources of finance and labour through ethnic ties and/or shared cultural values and language use (Vershinina et al. 2011). In examining access to different forms of capital, finance and opportunity, an important development in examining migrant entrepreneurship has been the emergence of a mixed embeddedness approach to understand migrant forms of entrepreneurship (Ram et al. 2008; Kloosterman 2010; Vershinina et al. 2011). This approach takes into account the embeddedness of migrant entrepreneurs within co-ethnic networks and seeks to contextualise them within broader social, political and economic spaces within the host country (Ram et al. 2008). It focuses on the many difficulties that migrant entrepreneurs face in the host country (access to finance, access to training) as necessarily part of the social and political context of the host country (Ram et al. 2008). As such, the mixed embeddedness approach argues for the need to focus not only on ethnic strategies but also personal strategies within specific opportunity structures, markets and regulatory environments.
Human capital and migrant entrepreneurship
The key constructs of human and social capital will have an impact on not only indigenous entrepreneurs, but also crucially for migrant entrepreneurs (Bagwell 2015), including those who are forced to migrate. Human capital consists of characteristics, skills, competences, education etc. that are inextricably tied up with the individual and that have a positive impact on him/her (Becker 1975). While links between human capital and entrepreneurship are the subject of debate, it has been suggested that there is a positive relationship between more formal human capital in the shape of higher educational attainment and entrepreneurship due to greater skills associated with opportunity recognition (Williams et al. 2017). Furthermore, more highly educated individuals are expected to have higher ability and a superior information set, and therefore more start their own businesses (Constant and Zimmerman 2006). As Wauters and Lambrecht (2008) state the process for leaving one’s home to start a new life in a new country is self-selective, with immigrants often being more prepared to take risks than those who stay at home, yet this does not apply to forced migrants as they have to leave their country, often quite suddenly, and thus lose access to social capital. Among these are also older people and people with country-specific skills and qualifications who in normal circumstances would not leave their home country (Gold 1988). For such migrants, over time, the stock of human capital acquired in the COR may better fulfill the needs of the host-country labour market (Bean et al. 2004). This means that as more human capital is acquired in the COR, it may increase the opportunities for salaried employment, rather than entrepreneurship (Donato et al. 2008). Given this discussion, we propose the following hypothesis to examine the role of human capital in migrant entrepreneurship:
-
Hypothesis 1: Migrants with higher level human capital: 2a) education before migration) 2b) participated in any business related training in COR; experience (years of migration) are more likely to be engaged in entrepreneurial activity as compared to their counterparts
Social capital and migrant entrepreneurship
Within the entrepreneurship literature, social capital is often seen as a resource, enabling migrant entrepreneurs to access co-ethnic social networks in order to gain finance (Vershinina et al. 2011). One of the most important contributions linking social capital to knowledge networks is that of Nahapiet and Ghoshal (1998), which focuses on the importance of social capital and the advantages and intellectual capital creation it facilitates through personal relationships which fulfill social motives such as sociability (Portes 1998), approval and prestige. Within this, the structural dimension refers to the interactions, network connections, patterns and strength of ties; the cognitive dimension considers shared vision, purpose and meaning, normally as a result of the shared history in families; while the relational dimension regards to trust, norms, obligations and identity, and also the quality of the connections. Social capital depends in the context; hence, higher levels of social capital indicate mutual interdependence between actors (Nahapiet and Ghoshal 1998). This work is particularly useful, not only because it makes the link between intra-organisational networks, knowledge and social capital, but because it focuses on the combination and exchange of knowledge in relation to factors such as access, motivation, capability and anticipation of value (Williams et al. 2017).
Most commonly, social capital consists of the perceived value inherent in networks and relationships generated through socialisation and sociability as a form of social support (Borgatti and Foster 2003). The network space can be defined by the nature of the relationships, interactions and ties underlying particular networks (Williams et al. 2017). For example, Putnam’s (2000) analysis of social capital presents an intricate examination of the types of network space occupied by actors across a range of interpersonal networks. The network can be perceived as the sum of the invisible links between nodes that form the network, and as such, these links provide the relationship and channels of communication between individuals (Jack 2005). However, these links are not simply relationships and channels of communication, they also act as the mechanism through which information and resources are drawn from, and exchanged with, the social context and provide the vehicle through which an individual learns about the social context and the social context learns about an individual (Jack and Anderson 2002). They also provide the bridges to new information and resources in other social structures and bond the various relationships between nodes together (Granovetter 1973) and are the key to unlocking and accessing the social capital that resides within a network (Anderson et al. 2007). Social capital’s power is its ability to understand how individuals are able to mobilize their network to enhance personal returns usually within place-bound environments (Westlund and Bolton 2003).
Social capital is a product of aggregated resources held within durable networks (Ram et al. 2017). While migrant networks vary in terms of the number of actors they incorporate, their location, the benefits they provide to individual members and the strength of relations between network actors can all influence economic activity (Smallbone et al. 2010; Lajqi and Krasniqi 2017). Indeed, networks offer migrant entrepreneurs invaluable and unique competitive advantages, making them fundamental to the growth and success of their own businesses, and a potential resource for harnessing entrepreneurship (Nielsen and Riddle 2010). Yet, the social networks that forced migrants can access are likely to be less extensive than that of voluntary economic migrants, who are often better able to access social capital through ethnic ties and/or shared cultural values (Vershinina et al. 2011). Due to the forced and sudden nature of their flight, they can lack networks in their COR, with the only option being to develop them over time. As they have experienced traumatic events both in their country of origin (COO) and during flight which can cause psychological problems which hamper self-reliance and entrepreneurship (Hauff and Vaglum 1993). Also, as a result of the nature of conflict, they are often unable to return home to acquire funds, capital or labour for their ventures which would supplement their limited social capital in the COR (Wauters and Lambrecht 2008). In order to examine social capital and migrant entrepreneurship, we propose the following hypothesis:
-
Hypothesis 2: There is a positive relationship between higher social capital: 1) fluent in COR language 1b) foreign spouse, 1c) co-ethnic networking of migrants, 1d) family size and probability of being an entrepreneur
Forced migration and entrepreneurship
While the economic impacts of voluntary labour migration have been examined, the literature on the development impacts of forced migration needs to be deepened (World Bank 2009). In recent years, forced migration due to conflict has reached its highest level since World War II, affecting more than 51 million people around the world (World Bank 2014). While forced displacement is often viewed as a humanitarian issue, given that it has impacts on growth, employment and public spending for countries of both origin and destination, it is also a major development issue. The majority of forced displacement is driven by conflict, generalized violence and/or human rights violations (UNHCR 2014). Palestinian refugees continue to comprise the biggest refugee group worldwide, with other major source countries for refugees being Afghanistan, Syria, Somalia, Sudan, the Democratic Republic of the Congo and the Central African Republic, while further displacements have been seen through conflict in Iraq and Ukraine (UNHCR 2014). Many forced migrants are in protracted situations, with estimates that two-thirds of refugees have been in exile for more than 5 years (World Bank 2014).
In the COR, the influx of displaced people impacts on growth and public policy aimed at integrating new populations and providing economic opportunities (Wauters and Lambrecht 2008). Increased labour supply can impact on wages depending on whether the forced migrants complement or substitute the skills of local workers, while growth is enhanced if migrants bring new skills and additional sources of financing, and increase trade and domestic demand (Enghoff et al. 2010). At the same time, an influx of forced migrants can place pressure on public services, and to counterbalance these effects, the COR needs to increase public expenditure, while at the same time developing policies to enhance social cohesion (World Bank 2014). For the forced migrants themselves, displacement has impacts on their human and social capital, income and employment, as well as entrepreneurial opportunity (Jacobsen 2005). Because of the nature of the displacement, forced migrants face significant loss of assets and networks, and this differentiates them from economic migrants (Ibanez and Moya 2009).
A key challenge of public policy is to find resilient solutions to the increasing number of refugees in protracted displacement situations (World Bank 2014). This means that policy makers need to examine not only the vulnerabilities of forced migrants, but also ways in which the COR can make the most of their education, skills and experience. One element of this which is currently under-researched is the role that entrepreneurship can play. A key unexplored question regarding forced migrants is the ways in which their resources may be harnessed in order to develop entrepreneurial ventures which have the potential to benefit the economy of the COR. We contend that forced migrants face acute occupational choice constraints. In the context of forced migrants, it can be argued that they were not ‘prepared’ for migration as they were not responding to opportunity, for example they had not secured employment before migration and/or were not responding to start a business. Also, the level of skills they acquired in their COO may not be adequate to respond the needs of labour market in Western countries. In such circumstances, migrants do not have many employment options and may be pushed to entrepreneurial activities. Accordingly, this discussion leads as to following hypothesis:
-
Hypothesis 3: Forced migrants are more likely to be entrepreneurs compared to non forced migrants
Empirical focus
The breakdown of the communist regime in the early 1990s was the beginning of the end of Yugoslavia and led to significant economic and political change across the Balkans. Several unsuccessful military campaigns to unite Serbs in neighbouring republics into a ‘Greater Serbia’ resulted in wars in which thousands were killed and many more became refugees. In Kosovo, some 800,000 people out of a population of 1.8 million were forced to flee the country, another 400,000 were forced to flee their homes and hide within Kosovo; 10,000 people were killed, and 3000 were abducted (Kellezi et al. 2009). Following NATO intervention and the conclusion of the wars, the former Yugoslavia was split into different nations, including Serbia which contained the formerly autonomous province of Kosovo. However, following the continuation of political and ethnic tensions, Kosovo unilaterally declared its independence in 2008 (Yannis 2009; Ivlevs and King 2012). It remains one of the poorest states in Europe (Ivlevs and King 2012; Krasniqi and Mustafa 2016; Williams and Vorley 2017) and has been plagued by high unemployment, low levels of growth, high levels of poverty and poor infrastructure (Krasniqi 2007; Hoxha 2009; Krasniqi 2014).
Kosovo has experienced four key waves of migration (Riinvest 2007; Vathi and Black 2007; UNDP 2012, 2014). During the 1960s, relatively unskilled (mainly) men from rural areas of the country emigrated to Germany and Switzerland as ‘guest workers’, and during this wave, those who emigrated sent money home which was often used to invest in houses. This migration flow of Kosovar Albanians was due to Yugoslavia’s move to liberal economic reforms in 1965, which included the opening of borders (King et al. 2010). Kosovo’s economic and political status under the former Yugoslavia was characterized by low industrial development and high unemployment, and as a result, substantial Kosovan migration also took place to other parts of Yugoslavia at this time. The flow of migration decreased briefly in the 1970s when new jobs were created in the public sector and socially owned enterprises in Kosovo. At that stage, Kosovo gained autonomous political status within the former Yugoslavia (UNDP 2014). However, even with the improved political status of Kosovo under Yugoslavia, there was still a high unemployment push for migration for the remaining working age population which was not able to be absorbed by the public and/or social enterprises sector. In the 1970s, the share of Albanians in the working age population was 68% while the share of the employed labour force was only 55%, and the share of Serbs in the working age population was 23%, whilst the share of employed labour was 33% (Bartlett 1990). As is clear, there was discrimination on ethnic grounds in the Yugoslav labour market.
In 1989, Kosovo’s autonomy was abolished and more than 150,000 ethnically Albanian workers were dismissed from the public service and socially owned enterprises pushing them either towards entrepreneurship in Kosovo or to migrate (Krasniqi 2012). Skilled and educated young men from both rural and urban areas migrated to Western European countries to find jobs and to escape from Yugoslav military service. Migration was seen as a means to escape poverty and improve the quality of life for family members left behind through remittances (UNDP 2014). When war broke out in 1998, many people were forced to migrate, mostly to Albania, Western Europe and the United States, and then when the war ended in 1999, Kosovo experienced a rapid return of the displaced population (UNDP 2014). Following the war, while progress was made towards political stability in Kosovo, immigration policies of Western European countries became less favourable for Kosovars. As a consequence, emigration waves have mainly consisted of migration for family reunification purposes, irregular migration of unskilled young people and (temporary) legal migration of highly skilled and highly educated individuals through study or work arrangements (UNDP 2014).
The latest figures estimate that Kosovo’s diaspora is estimated at approximately 700,000 people, equivalent to 40% of the resident population (UNDP 2014). Kosovo’s diaspora reside in the some of the most highly industrialised Western economies, where they are able to accumulate substantial capital, gain advanced skills and develop effective networks. The largest shares of Kosovar migrants are estimated to be in Germany (33% of the diaspora population) and Switzerland (25%), followed by Italy (8%), Slovenia (6%), Austria (6%) and the USA (4%), with the remainder mostly in Scandinavian and Benelux countries (Kosovo Agency of Statistics 2013). Approximately 95% of the household heads in these countries are employed (Office of the Prime Minister 2016), although figures on the proportion of those who are entrepreneurs are not available.
Research design and data
The study uses Riinvest Migrant’s Survey data collected at the end of 2008 and beginning of 2009.
This study is supported financially by Foundation for Open Society in Kosovo, Balkan Trust for Democracy, Friedrich Ebert Stiftung. One of the authors is Riinvest research affiliate and has been working for Riinvest for several years, therefore was part of the team and had access to data. The data were collected by trained enumerators. The data is based on a large survey of 715 (few observation were deleted for inconsistency or missing observations) migrants and contains detailed information about the individual characteristics (e.g. gender, age, education, occupation status, income [if working] and migration history), household characteristics (e.g. household size, marital status and share of employed family members). The aim of the Riinvest’s survey was to investigate the profile, characteristics and the potential of Kosovan diaspora for development of the country. The unit of analysis was individual migrant living abroad. The details on interviews were confidential—fully anonymised.
We use intercept point sampling with location sampling or aggregation point sampling (McKenzie and Mistiaen 2009; Krasniqi and Williams 2017) which have been used in other similar migration studies in USA and Mexico (e.g. Wasserman et al. 2005; Bustamante et al. 1998). There is no exact size of our population frame of Kosovan diaspora, but the intercept point sampling ensures generalisability of results and it is first best method to track migrants. The intercept point sampling design included interviews with migrants in border crossing points of Kosovan Diaspora (Prishtina International Airport, Hani i Elezit, Merdare, Porti i Durrsit in Albania, Gjilan). Multiple border locations were chosen for better coverage of the population of interest, as suggested by McKenzie and Mistiaen (2009). These multiple crossing points are the main entrance routes to Kosovo for more than 90% of migrants. Systematic sample is employed to select random individuals crossing the border at the specific the location during the specified time periods during the day.
The advantage of our survey is that it was undertaken during December 2008 and January 2009, which marked a major flow of migrants to Kosovo, especially considering that in 2008 the country declared its independence and the migration flow (both to Kosovo and externally) was at its peak. This allows better representation of intercepting migrants by allowing us to survey individuals who are seldom found in Kosovo. This sampling design is feasible in cases where the sampling frame is absent, because it is impossible to have a list of the total number of migrants (see McKenzie and Mistiaen 2009; Gedeshi and De Zwager 2012; Krasniqi and Williams 2017, for similar approach) and is much closer to random sampling (McKenzie and Mistiaen 2009). Although our study uses data from 2008/2009, the results and hence conclusions are coherent with the current situation of migration and their entrepreneurial activity abroad, mainly in EU. Moreover, the results are comparable, and recommendations may be applied to the other context in terms of entrepreneurial activity of migrants in EU. In addition, the theories we test—mainly human capital and social capital—imply the use of variables that are not sensitive to time, such as foreign spouse, foreign language, networking, forced migration.
Model
To analyze the relationship between the individual and socio-demographic and entrepreneurial of Kosovan diaspora, we use a series of binomial logit regression models, where,
$$ {y}_i=\left\{{}_{0_{if\kern0.5em i\kern0.5em is\kern0.5em employed}}^{1_{if\kern0.5em i\kern0.5em is\kern0.5em self- employed}}\right. $$
The binomial logit takes the following general form:
$$ {y}_i={X}_i\beta +\varepsilon $$
where yi is the dependent variable indicating whether the individual migrant is an entrepreneur and own a company, and Xi is the vector of explanatory variables (individual and socio-demographic level variables). The description of explanatory variables used in the regression analysis is presented in Table 1 in the Appendix.
Table 3
Correlation matrix of explanatory variables
Variables | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | Gender (male = 1) | 1.000 | ||||||||||||
2 | Migrant’s age | 0.183 | 1.000 | |||||||||||
3 | Migration duration (years) | 0.060 | 0.639 | 1.000 | ||||||||||
4 | Marital status (married = 1) | 0.094 | 0.353 | 0.152 | 1.000 | |||||||||
5 | Employed family members | 0.042 | 0.364 | 0.302 | 0.292 | 1.000 | ||||||||
6 | Fluent in COR language | − 0.090 | 0.203 | 0.271 | 0.027 | 0.141 | 1.000 | |||||||
7 | University education prior migration | − 0.132 | − 0.149 | − 0.208 | − 0.017 | − 0.080 | 0.104 | 1.000 | ||||||
8 | Business training in COR | − 0.040 | − 0.013 | − 0.021 | 0.019 | − 0.003 | 0.052 | 0.010 | 1.000 | |||||
9 | Co-ethnic networking in COR | − 0.076 | 0.011 | 0.079 | − 0.039 | 0.015 | 0.040 | − 0.011 | − 0.032 | 1.000 | ||||
10 | Forced migration (Kosovo War, 1999 = 1) | 0.008 | − 0.008 | − 0.092 | − 0.033 | − 0.037 | 0.022 | − 0.016 | − 0.021 | − 0.005 | 1.000 | |||
11 | Foreign Spouse | 0.018 | − 0.083 | − 0.083 | 0.178 | − 0.004 | 0.186 | 0.115 | 0.003 | − 0.011 | − 0.004 | 1.000 | ||
12 | Family size | 0.014 | 0.405 | 0.282 | 0.521 | 0.640 | 0.093 | − 0.084 | 0.003 | − 0.018 | − 0.025 | 0.011 | 1.000 | |
13 | Secured job in COR before migration | − 0.006 | 0.064 | 0.009 | 0.027 | 0.076 | 0.040 | 0.124 | 0.027 | 0.059 | 0.015 | 0.061 | 0.029 | 1.000 |
Results
To evaluate the effect of personal, socio-demographic and networking variables on the probability of entrepreneurship among Kosovan migrants, we here report the results from a series of logit regression models. The first regression in the analysis estimated a baseline model with individual level variables including gender, age, marital status, employment of family members, university education before migration, business training in host country. In addition, we have included variables to measure the connection to diaspora, whether the migrant was forced to migrate or not, if migrant has a foreign spouse and proficiency in host country language, as this can have an impact on entrepreneurial activity (Wauters and Lambrecht 2006, 2008). These variables are intended to capture the effect of acculturation and adaption to host country. We have also included some control variables such as years of migration, family size and dummy variables if migrant has had secured a job before migration. All regression models passed the Wald test, indicating that all independent variables are jointly and statistically significant and different from zero. Correlation matrix reported in the Appendix concludes that multicollinearity is not a problem in our data, as correlation coefficients are generally low. All maximum likelihood models have passed the statistical test for joint statistical significance of explanatory variables (see Wooldridge 2005).
Model 1 in Table 2 in the Appendix reports that the probability of being entrepreneur is increased with age, if the migrant is male, is fluent in the language of the host country, and has been forced to migrate, when taking into account and holding constant other characteristics, and is line with Wauters and Lambrecht’s (2006, 2008) findings that such capitals can influence the activity of forced migrants. This reveals that male migrants have a higher likelihood than women to be an entrepreneur, which is similar to earlier studies in other global regions (Caliendo et al. 2014b; Caliendo et al. 2014a; Krasniqi 2009). Findings suggest that connection to their ethnic group (denoting co-ethnic networks) has no effect on the probability of being an entrepreneur. This effect may be compensated by the other variables that a found to be highly statistically significant such as connection and networking to host country network and acculturation. This is in line with previous studies suggesting that contacts and networking with native population are critical, referred to in the literature as an element of bridging social capital (Putnam 2007), and contrasts with other studies which suggest that ethnic co-networks may help with access to resources and information in helping their members to set up their own business, using previous experience acquired by the established immigrant communities (Levie 2007; Masurel et al. 2004; Jones et al. 2010).
The variable ‘fluent in foreign language’ is statistically significant suggesting that individuals are socialised and well networked with native population. This may increase the chances of spotting and developing entrepreneurial opportunities and getting help from native friends, as well as benefitting from their network. This finding is reinforced in model 2 when introducing the variable foreign spouse, which is found to be highly significant and positive. Being married to a native of the COR exposes migrants to better access to information and networks and of course may have better legal status as migrants. This study is similar to other studies that found that married immigrants are more likely to go into entrepreneurship (Constant and Zimmermann 2006). According to Constant and Zimmermann (2006), marital status plays a role in entrepreneurial activity in two ways. First, marriage brings stability, and married men, it has been shown, are more productive through the division of labour and household production, which in this case can increase their likelihood of being business owner. Second, married men can benefit from their wife’s support and can count on her helping with their business as well. Being married can act as a risk-sharing phenomenon and married men who are entrepreneurs can also count on their wife’s stable income, if she is working, while enabling him to use his income for business (Krasniqi 2009).
To control for differences in the initial stock of human capital, we use the variable education before migration in home country and variable denoting any specific training in destination country. We include these two separate variables in regression. We find no evidence of education of the migrant before migration, suggesting that the quality of education in Kosovo is not adequate and supportive for building entrepreneurial skills necessary for start-ups. Considering that diaspora in our sample is usually located in EU Countries and USA, this may suggest that education in Kosovo is not in line with requirements of developed market economies. While origin-country human capital does not necessarily restrict salaried employment chances, because they usually do low skilled work with few educational requirements, it may constrain the probability of being an entrepreneur. As Kanas et al. (2009) argue, migrants entering entrepreneurship need access and ability to process large and complex business-related information (e.g. market size, consumer products or reliable suppliers) which is country specific. This supports our finding that having a foreign spouse and being fluent in the language of the COR may compensate for a lack of skills and information needed for start-up. In addition, our analysis finds that securing a job prior to migration influences the likelihood of becoming an entrepreneur. This may be because this first job provides them access to a wider network, often outside of their own ethnicity, which they can later exploit to launch an entrepreneurial venture.
On the contrary, any additional training (variable business training in model 2) migrants received in those countries seems to play a significant and positive impact on probability of being a business owner. Constant and Zimmermann (2006) maintain that better educated individuals are expected to have higher ability and allow individuals to know more and to have superior information set, we would expect that more years of schooling will push individuals into entrepreneurship. Those with education and work experience acquired in the COR will tend to better fulfill the needs of the host-country labour market (Bean et al. 2004). Therefore, the stock of human capital accumulated in the host country tends to increase the opportunities for salaried employment and decrease the likelihood of entrepreneurship (Bean et al. 2004; Donato et al. 2008). Migration experience is another important factor determining the probability of being a business owner. Specifically, we create the variable to measure migrant’s exposure to the host country business environment as well as accumulated labour market experience. The variable years in migration suggest that there is an inverted U-shaped relationship between probability of being a business owner and year in migration. In our model, the turning point where the effect of years in migration of entrepreneur becomes zero is 14.38 years.1
Finally, we did not find evidence that employed family members and family size have impact on entrepreneurial activity of migrants. The findings contrast with a previous study on immigrant entrepreneurs in Sweden revealing that family resources are important for immigrants to integrate economically into a country (Bird and Wennberg 2016). We find that for migrant entrepreneurs, it is more important to capitalise from benefits of marriage with a native of the COR than from family resources (family size and employed family members).
Discussion
The contribution of our paper is to show how migrants harness their human and social capital resources in order to undertake entrepreneurial activity. The analysis shows that the probability of being entrepreneur is increased with age, if the migrant is male, is fluent in the language of the host country and has been forced to migrate, when taking into account and holding constant other characteristics. It demonstrates that male migrants have a higher likelihood than women to be an entrepreneur. In addition, connection to their ethnic group (denoting co-ethnic networks gained through social capital) has no effect on the probability of being an entrepreneur, suggesting that bridging social capital is limited in terms of entrepreneurial development (Williams et al. 2017). This effect may be compensated by the other variables that are found to be highly statistically significant such as connection and networking to the host country network and acculturation, meaning that social capital can be accessed across international borders. This contrasts with other studies which suggest that diaspora co-networks may assist in accessing social capital which help migrants to set up their own business previous experience acquired by the established immigrant communities (Levie 2007; Masurel et al. 2004; Jones et al. 2010). Being fluent in the foreign language of the COR is statistically significant, suggesting that human capital is enhanced through language skills, and that social capital is developed as individuals become more socialized and networked with the native population, and may increase the chances of identifying and developing entrepreneurial opportunities and getting help from native friends. Human capital associated with fluency in the foreign language represents higher human capital, but also that they are able to access a wider network of social capital resources in the COR.
Furthermore, being married to a native of the COR exposes migrants to better access to bridging social capital, and of course may have better legal status as migrants. Marital status means that married males can benefit from their wife’s support, and being married can act as a risk-sharing phenomenon as one partner’s income can be used to develop the entrepreneurial activity of the other. Here, the work of Nahapiet and Ghoshal (1998) is instructive as the structural, cognitive and relational dimensions of social capital can have an important impact on activity. We add to this literature by showing that access to wider social capital through marriage to a native of the COR can create more collective trust and shared knowledge, which benefits the migrant as well as the wider economy if those networks can be effectively utilised. Given that social capital depends on the context in which it is gained, higher levels indicate mutual interdependence between actors (Nahapiet and Ghoshal 1998), as marital status and family links not only increase capital endowments, but also act as a risk-sharing phenomenon.
Migrant entrepreneurs often take on transnational characteristics, which are a mix of learned cultural and social behaviours from their country of origin and host country (Kshetri 2013) and which give them advantages in conducting entrepreneurship (Bagwell 2015). Yet, our findings demonstrate that these advantages are deferred as human and social capital is developed by the forced migrants. Indeed, we show that education of the migrant before migration is not significant, suggesting that the quality of education in the COO, in this case Kosovo, is not adequate and supportive for building entrepreneurial skills necessary for start-ups. However, any additional business training that migrants gain in their COR seems to play a significant and positive impact on the probability of being an entrepreneur, showing that such human and social capital is acquired over time and impacts on entrepreneurial activity. This demonstrates the lack of ‘preparedness’ of forced migrants as they are not moving to exploit an opportunity which utilises their skills; rather, they move involuntarily as a result of conflict and then need to acquire skills for them to exploit entrepreneurial opportunity.
Conclusions
The key research question for this study was ‘How do human and social capital resources influence the entrepreneurial activity of migrants, with special reference to migrants who have been forced to migrate?’ In the above discussion, we show the role of human and social capital in fostering entrepreneurial activity among forced migrants. The capital endowments of forced migrants can be weak, characterised by low levels of human capital and limited social capital access in the COR. However, through education and training, as well as access to wider networks, social capital can be built up over time and plays an influence in fostering entrepreneurial activity.
The results of our study are important considering that migration is an ever more important phenomenon, with millions of people moving around the world either in response to better economic opportunities outside their COO or in response to some form of crisis (World Bank 2014). These waves of migrants are not simply a humanitarian issue, but also an issue of economic development both in the COO and the COR. Our study focuses on how human and social capital endowments of forced migrants influence their likelihood to be an entrepreneur in their COR. Despite its growth as an area of study, the entrepreneurial activities of migrants do not seem to be prominent within ‘mainstream’ entrepreneurship research (Ram et al. 2017), and there is even less engagement with the topic of forced migration (Bagwell 2015).
As such, there is great scope to expand this understanding, and a key area which is under-researched is that of forced migration. Such migration, often by individuals who have to leave their COO suddenly, creates its own pressures on the COR, with increased labour supply impacting on wages depending on whether the forced migrants complement or substitute the skills of local workers, as well as the potential for enhancement of skills and other resources and increase trade and domestic demand (World Bank 2014). However, pressure is also placed on public services, and the COR needs to respond effectively. For the forced migrants themselves, displacement has impacts on their human and social capital, income and employment, as well as entrepreneurial opportunity (Jacobsen 2005). Because of the nature of the displacement, forced migrants face significant loss of assets and networks, meaning that social capital built up over time in their COO can be lost. As policy makers seek new and more effective methods of integrating migrants and maximising their economic and social potential, exploring the potential of (forced) migrants to be entrepreneurs may be an effective avenue for policy support (Wauters and Lambrecht 2008).
With regards to policy implications, as with most aspects of entrepreneurship policy, there is no magic bullet (Huggins and Williams 2009, 2011). However, countries receiving large waves of forced migration should consider providing training for business start-ups as this seems to have a significant effect on preparing migrant to enter into entrepreneurial activity. As the World Bank (2014) has made clear the recently approved Sustainable Development Goals (SDGs) place pressure on countries to ensure safe, orderly and regular migration, limiting exploitation and abuse, and one path towards this is ensuring the entrepreneurial capacity of the migrants is harnessed. By doing so, these countries can compensate for the often inadequate entrepreneurial skills gained from the migrants’ COO, and thus increase their productive potential. The improvement of entrepreneurship curricula and embeddedness in school and university education should be a priority, so that it may assist in fostering the entrepreneurial activity of future migrants at home and abroad. This is particularly important considering that approximately 36% of recent refugees in Europe are children. Increasing their entrepreneurial skills while at school may not pay a short-term dividend, but over time can positively impact on their entrepreneurial potential and thus wider economic growth (World Bank 2014).
We acknowledge that the article contains some inherent limitations. Our study took place in 2008/2009 and focused on migrants returning to Kosovo. This limits the generalisability of our findings, as it does not consider migrants in situ within their host country. While the article focuses on the single country context of Kosovo, which again limits generalisability, it nonetheless provides lessons for other countries (both the COO and COR) where conflict has seen forced outward migration. Recent examples of Syria and Ukraine demonstrate how conflict creates large migration movements (World Bank 2014) and how countries often struggle to provide effective solutions, either on their own or part of supranational bodies. Future research is needed to examine how these waves of migration have an impact on development, and how the skills, experience and knowledge of forced migrants can be harnessed to benefit themselves, their country of residence and their country of origin.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.