Angela Parenti (p), Davide Fiaschi, Cristina Tealdi

Jacques Poot

This paper presents a new methodology to estimate the self-selection of migrants using
an intertemporal random utility maximisation model. This methodology, which
accounts for multiple destination choices grouped into nests, overcomes some of the
issues typically discussed in the literature, such as the lack of available data on migrants
and the presence of unobservable variables. We apply this methodology to
inter-regional migration in Italy in the period 1992-2016, and we find strong evidence
of self-selection among Italian regions on geographical base. Specifically, we
quantify the return to education to be higher in the best Northern regions compared
to the worst Southern regions by approximately 2 percentage points, which leads to
a comparative higher inflow of skilled workers in Northern regions. However, at the
same time, some Northern regions, as Lombardy and Piedmont, also show a relatively
high outflow of skilled workers in favour of other Northern regions such as
Emilia Romagna and Veneto.

Rodrigo Perez Silva (p), William Foster

Angela Parenti

The usual approach when studying migration is to describe either the characteristics of potential migrants that affect the likelihood of the migration decision, or the place-specific attributes that attract or repel migrants, or, as addressed in more recent literature, a combination of both. In the analysis of the role of the place-specific attributes of origins and destinations, the literature points to differences in unemployment, poverty, and income, among other factors, that influence population migration and relocation decisions. There are, however, different types of potential migrants who respond differently to different incentives and place-specific characteristics. This is especially likely to be true for lagging sub-regions within a sufficiently large primal-city area or country. This present paper aims to describe how different stages of development in a region of origin for emigrants determine migration decisions and whether different population sub-groups are more or less responsive at these different stages of development. Preliminary results suggest that, first, individuals are more likely to migrate when their own region’s development is either very low or very high. And that, second, conditional both on the attributes of the destination region and on individual characteristics, the level of economic development of the emigrant’s home region selectively affects the migration decision. In particular, early stages of economic development are associated with migration of relatively low-skilled workers, whereas the opposite is true at more advanced stages of development. Empirical results are presented for the case of internal migration in Chile, over the 2000-2017 period.

Holger Seibert (p)

Rodrigo Perez Silva

The German system of vocational education and training (VET) is very much firm based. Training takes place in private companies and is accompanied by vocational education in public vocational schools. Companies offer training positions according to their economic well-being as well as to future requirements in skilled manpower.
For many years the German VET market was not able to provide school leavers with the necessary number of VET vacancies - especially in East Germany and especially for foreign youth in West Germany. The problems in East Germany were rather economy driven, because smaller East German companies have always had lower resources to stem the necessary supply. In the most difficult years there have been almost twice as many applicants as open VET positions. Consequently, sufficient numbers of East German School leavers successfully applied for VET positions in West. In West Germany it was rather problematic for marginalized groups to find VET positions, such as school leavers with low achievements or foreign youth. As mobile VET applicants from East Germany usually had better school degrees there used to be much more competition for these marginalized groups.
Our paper analyzes if and to what extend the growing opportunities for foreign youth in West Germany are a result of the demographic changes in East Germany and the subsequent withdrawal of East German youth from the West German VET market. For this purpose, we use employment registry data from the Federal Employment Agency as well as population data from the Federal Statistical Office. Detailed analyzes prevail in how far apprentices with foreign nationality are to be found in VET positions that used to be occupied by applicants from East Germany in the past. Thereby, regional analyzes on NUTS-3-levels play an important role, as foreign population in Germany is concentrated in urban districts. Furthermore an occupational differentiation is necessary as foreign youth clusters in certain apprenticeship occupations. Thus, we compare regional occupation clusters where competition between East German and foreign youth took place.
At the same time it is vital to disentangle to what extend the improving opportunities for foreign youth are caused by the improved economic conditions in West Germany alone. Therefore, we consider a longer time period beginning already in 1999, when East German school leaver cohorts used to be much bigger and the East German VET market was rather consistently fierce.

Arthur Grimes, David Maré, Jacques Poot (p), Shaan Badenhorst

Holger Seibert

One of the main challenges facing non-metropolitan regions is the retention of highly-educated young people. A loss of the brightest and best can to lead to reduced business creation, innovation, growth and community wellbeing in such regions. In this paper we use rich longitudinal microdata from New Zealand’s integrated administrative data infrastructure to analyse the geography of the destination choice of tertiary graduates, conditional on prior school location. We address this question of post-student location choice in the context of the approach of Chen and Rosenthal who defined statistics measuring ‘quality of life’ (QL) and ‘quality of business’ (QB) for urban areas. Specifically we test whether students – of different characteristics (e.g. level and field of study, gender, age, ethnicity) – locate in places that are regarded as good to live or good to do business. By incorporating prior school location we also test how the pull of “home” is magnified or diminished by a location’s QL or QB. The results are important for local decision-makers who wish to attract and retain young qualified people to their area, and for decisions about the location of various types of tertiary educational institutions.