Jose Blanco-Álvarez (p), Manuel González-López

Carla Sá

Human capital has risen to prominence as the main driver of long-term economic development justifying massive public investments in higher education. However, the existence of externalities in the form of agglomeration economies can induce highly skilled individuals to migrate into a reduced number of dynamic urban areas hosting the lion’s share of knowledge-intensive activities.
Using a rich administrative dataset on labor contracts and business demographic statistics we build a gravity model explaining inter-regional migration by skill level between Spanish regions (Autonomous Communities or CCAA; equivalent to NUTS2) during the last 22 years (2001-2022).
Our estimations indicate that the agglomeration of knowledge-intensive activities (especially proxied as knowledge intensive business services, or KIBS) is a significant and strong determinant of high skilled migration after controlling for wage differences, the provision of amenities, housing prices or the existence of different languages among others. These results are robust to different specifications, including OLS, dynamic models or Pseudo-Poisson Maximum Likelihood estimations with different sets of fixed effects by region of origin, region of destination and year.
These results also pose an important challenge for policymakers, as they implied that efforts on increasing education opportunities at home regions might be seriously undermined by initial economic differences. There are indicatives of high human capital agglomeration inside other countries (for example London in the UK or Paris in France) and given that no (formal) barriers to migration exist at internal level, this case study is of relevance for them. To the extent that the EU is a free movement area, the case of Spanish regions might be also representative of dynamics taking place at upper levels.

Pedro Luís Silva, Carla Sá (p), Madalena Temudo

Giulia Bettin

This study examines the role of student mobility, particularly relocations resulting from not being admitted to a preferred institution, in shaping higher education efficiency and regional development. Portugal’s centralised admission system assigns students based on grades, preferences and institutional capacity, often leading to placement mismatches that make students relocate. Such forced mobility may contribute to inefficiencies, including higher dropout rates, delayed graduation, and suboptimal allocation of educational resources. By exploring the determinants and consequences of student mobility, this paper seeks to provide insights into the spatial distribution of human capital and its implications for regional growth and educational equity.

To analyse these patterns, we apply a gravity model of student flows, estimating the determinants of mobility between municipalities and higher education institutions. Using administrative data from the Portuguese Directorate-General of Higher Education, we model mobility as a function of first-choice allocation, geographical distance, institutional quality, and field-specific constraints. The gravity framework allows us to assess how institutional characteristics and regional economic factors influence student movements and whether specific fields exhibit higher mobility barriers.

By focusing on the relationship between mobility patterns and the structural features of the higher education system, this study contributes to the broader discussion on educational efficiency and spatial equity. Understanding the extent to which institutional constraints versus individual preferences drive student flows can inform policies to improve access, reduce mismatches, and foster balanced regional development. Our findings will help assess whether current admission mechanisms promote an optimal distribution of human capital or whether adjustments—such as targeted financial incentives, institutional specialisation strategies, or policy interventions to mitigate mobility costs—are needed to enhance the efficiency of higher education.

Giulia Bettin (p), Silvia Mattiozzi

Marcin Stonawski

Globalization had a profound impact on advanced economies, driving down the prices of consumer goods and production inputs while simultaneously contributing to job losses in certain sectors (Dorn and Levell, 2024). Individuals affected by plant closures and mass layoffs often face diminished re-employment opportunities and reduced earning prospects, alongside worsening social conditions. Beyond economic and social consequences, plant closures may also lead to demographic shifts, as individuals reassess their residential choices in response to economic distress, a pattern observed also in the aftermath of natural disasters or political upheavals.
Using a staggered difference-in-differences approach, we investigate how manufacturing decline affected the internal mobility of Italy’s working-age population across local labor markets (LLMs) from 2000 to 2019.
The analysis relies on ISTAT microdata tracking residence transfers among Italian municipalities, aggregated at the LLM level. Labor demand shocks are identified by focusing on LLMs designated as areas of complex industrial crises by the Ministry of Economic Development, ensuring that the study captures broader local economic effects rather than firm-specific impacts. The study also explores heterogeneity in migration responses based on individual characteristics — such as gender, citizenship, and age — as well as LLM attributes, including urbanization levels, industrial intensity, and structural conditions.
Preliminary findings indicate a decrease in net migration following industrial crises, due to both reduced inflows from other Italian LLMs and increased outflows towards other LLMs. The results remain robust across different model specifications, alternative DiD estimators, and extended pre-treatment periods. Both Italian and non-Italian citizens exhibit lower net migration, though the effect is apparently driven by increased outflows among foreign citizens. No significant gender differences are observed in migration propensity. Additionally, district-based local labor markets experience a net population decline twice as large as non-district local labor markets. The demographic response to industrial crises is significant in the Center-North of the country but not in the South.

Marcin Stonawski (p), Vegard Skirbekk

Jose Blanco-Álvarez

In the contemporary world, there is a megatrend towards automation and migration. In Europe, the factors of depopulation, ageing populations and high levels of education are driving immigration to countries within the region.

Migrants from outside Europe predominantly occupy positions the labour market that are susceptible to replacement as a result of automation in the near future. Consequently, these individuals are at a heightened risk of unemployment and may inadvertently impose an additional burden on the host society, rather than providing a solution to the issue. It is therefore important to ascertain whether this elevated individual risk is mitigated at the level of the household. For example, the consequences of losing one's job may be less severe if one's partner is employed in a secure position, and more significant if one is the primary income provider with numerous dependents.

The present study focuses on evaluation of the risk of technological-induced job loss among migrants, based on indiviual data from Danish registers and RTI methology. The data allow us to assess both individual and household vulnerability to technological change and create a typology of households in this respect. The preliminary findings indicate considerable diversity among individuals and households with regard to the risk of job loss.