Alexandra Espinosa (p), Luís Horna, Rodrigo Mendieta-Muñoz, Nicola Pontarollo

Viktor Venhorst

This paper shows that the threshold model appropriately analysed can explain the emigration process as a consequence of the network features, where the willingness to emigrate depends on the personal and household conditions borne by the emigrant. In such a case, the emigration process can be defined in terms of a Pólya schema, and the emigration rate converges almost surely to a random variable with a beta distribution, which fully characterises the network (assimilation, integration, separation or marginalization). The model is applied to Ecuador, which has experienced an unprecedented wave of emigration since 1998. We show that, despite the massive exodus, the process does not constitute a diaspora, because Ecuadorians do not appropriately interact with natives.

In this paper we model the four strategies identified by Berry (1997), constructing a framework that combines the properties of the threshold model and the network framework to find the statistical properties of the network of migrants. Our main contribution is the characterisation of the strategies (assimilation, integration, separation and marginalisation) through the statistical properties of the emigration rate. This approach can be considered an analysis prior to any intervention, predictive for countries that do not have experience of integration of immigrants from a specific origin. The paper is structured as follows. In Section 2, we introduce integration and migration theories and show that the sequence of decisions (to emigrate or not) can be modelled as a Pólya Urn, under the assumption that the probability of emigration is a continuous function of the emigration rate. Therefore, we show the emigration rate has a beta limiting distribution, which fully characterises the network created abroad by migrants of culturally homogeneous groups. In Section 3, we apply this framework to the Ecuadorian case. Section 4 provides some conclusions and policy implications.

Anna D'Ambrosio (p)

Alexandra Espinosa

A growing literature studies migrants' contribution to the internationalization of regions (Gould, 1994; Rauch and Trinidade, 2002; Wagner et al., 2002; Felbermayr et al., 2015; Burchardi et al., 2018). Migrants are supposed to promote trade by facilitating the flow of information between the trade partners and by sustaining the enforcement of contracts.
However, only a few studies (Peri and Requena-Silvente, 2010; Briant et al., 2014; Burchardi et al., 2018; Bratti et al., 2018) sought to disentangle the mechanism underlying such
effect while undertaking a sub-national focus. Drawing on rich transaction-level data on Italian exports provided by the ISTAT, we seek to dig deeper into the understanding of
the mechanism underlying migrants' pro-trade effect, focusing on Italian NUTS3 regions.
Transaction-level data allow identifying differential effects of migration on the number of transactions vis-a-vis the average value per transaction; they allow aggregating transactions
according to the level of differentiation of the goods exchanged, which should be informative about the informational content of the good (Rauch, 2001; Rauch and Trinidade, 2002); and they allow aggregating transactions by technological intensity of the sectors of the trading firm.
We employ a gravity model of international trade (Anderson and van Wincoop, 2003; Head and Mayer, 2014; Chaney, 2008) augmented with migration where we allow for differentiated exporting capacity of Italian provinces (Bratti et al., 2018; Burchardi et al., 2018), differentiated importing capacity of the partner countries, and for heterogeneity in trading at the region (NUTS2)-country level.
We document a robust effect of both immigrants and emigrants on trade value and on the number of transactions, but not on the average transaction value. The results are consistent with the idea that migrants bridge the fixed, rather than the variable costs of trade (Chaney, 2008; Peri and Requena-Silvente, 2010). The effect is stronger for differentiated goods and for lower-tech sectors. Overall, the results suggest that migrants' effect is primarily an information effect that bridges the realization of new trade relationships, rather than increasing their value, which applies to sectors of relatively accessible informational content.
Moreover, our results reveal that migrants' effects correlate with the province exporting capacity, consistent with Bratti et al. (2018), suggesting a role for migrants in exports that
goes beyond the effect on bilateral trade costs onto the productivity of the firms in the region.

Titissari Titissari (p), Philip McCann, Niels Hermes, Viktor Venhorst

Anna D'Ambrosio

This paper aims to analyse how financial behavior becomes part of migration behavior by utilising the case of interregional migration, domestic remittance, and individual financial attitudes of over 30,000 individual sample in Indonesia.
The number of interregional migrants has outpaced the number of international migrants from Indonesia (UNDESA, 2017). This significant part of Indonesian internal migrant population has generated a growing flow of migrants’ remittances toward families who have remained behind. Our paper draws attention to see how financial behavior becomes part of migration behavior. We aim to examine the effect of interregional remittances to an individual's attitudes toward finance. Although there is a growing body of literature address international remittance and its effects of international remittance on consequent financial inclusion at the micro-level, previous studies of remittance have not dealt with the intra-country migration (internal migration across desa/village, district of kecamatan, kabupaten, province), interregional-remittance flow, and their consequences on the individual financial inclusion. Migrant remittances may foster access to and usage of financial services for the remittance-receiving individuals who tend to be characterized as the unbanked people.
Using the uniquely built data set drawn from the Indonesian Family Life Survey (RAND, 2014), we investigate whether or not remittance influences individuals to be more active in their financial lives, for example, in term of saving attitude. We conduct three stages of analysis: cohorts’ descriptive analysis using K-Means cluster, internal migration determinant analysis using logistic regression, individual saving attitudes analysis using multilevel logistic regression.
Since we aim to analyze how financial behavior becomes part of migration behavior among different cohorts, it is a prerequisite to analyze the migrants' behavior and factors determining their movement. Our descriptive analysis shows a pattern in which people who migrate, sending remittance, and does not have any saving. Therefore, we run further analysis to meet our ultimate research goal that links remittance and individual financial behavior.

Viktor Venhorst (p)

Titissari

In attempts to stimulate entrepreneurship in a general sense and university-business relations specifically, entrepreneurship is given more and more emphasis in curricula of Institutes of Higher Education (HEIs). Students are taught and stimulated to start their own businesses early in their careers. Unfortunately however, little is known about the impact of student self-employment on the development of their careers later on, either in self-employment or in wage employment. This study assesses, using longitudinal Dutch register data on recent HEI graduates, to what extent self-employment experience as a student influences the propensity of being self-employed after graduation, earnings from self-employment after graduation and finally, wages from dependent employment after graduation.

We find that, on average, self-employed students outperform their peers without self-employment experience: In the first years after graduation, they earn higher wages and, if running a firm, have higher income from self-employment.

After controlling for selection into student self-employment, the enabling effects of partners and differences in industries, we find positive effects on continued self-employment for men and women, as well as on the likelihood of being in dependent employment for men. However, we find no effect of self-employment experience as a student on subsequent income from self-employment. Also, we find evidence for slight negative effects on wages further into the careers of men and women.

This leads us to conclude that, even though student self-employment signals persons with great potential on the labour market, it does not foster more successful outcomes. We find no convincing evidence of a learning effect that can be capitalized upon in the early labour market careers of graduates.