Dr. Luca Alfieri (p), Prof. Vicente Royuela, Prof. Ilaria Mariotti, Dr. Federica Rossi

This paper examines the relationship between flexible working arrangements and residential location choices in Europe using data from the MOBI-TWIN survey of approximately 9,000 respondents across EU countries. We employ a two-stage modeling approach: first-stage probit models estimate relocation intentions within a five-year horizon, while second-stage multinomial probit models analyze destination preferences among capital regions, larger cities, smaller towns, and rural villages. Results show that flexible working arrangements significantly increase relocation intentions, controlling for individual socio-economic characteristics. However, the presence of children, life satisfaction in the current place of residence, and commute time from current residence to workplace reduce mobility propensity. Second-stage models reveal that flexible work support availability in potential destination communities is negatively associated with rural village preference. Our findings suggest that flexible work increases geographic mobility but does not uniformly promote non-urban relocation, challenging assumptions about remote work enabling rural resettlement. These results highlight the complex interplay between work flexibility and location preferences, with important implications for European policies addressing regional imbalances.

Patrick Bigler, Ms Marie Breuillé (p), Nicolas Chesaux, Camille Grivault, Sebastien Houde

This paper examines residential migration between rural and urban areas in France using a structural gravity framework applied to exhaustive inter-municipal (EPCI-level) migration flows from the French population census over 2012–2020. We estimate Poisson Pseudo-Maximum Likelihood models with origin-year, destination-year and dyadic fixed effects to assess the relative importance of spatial frictions, economic conditions, housing markets and environmental risks in shaping territorial attractiveness. We also explore heterogeneity between rural and urban territories and compare pre- and post-Covid patterns to evaluate whether recent shocks altered established mobility structures.

Our results reveal a very strong degree of spatial inertia. Once high-dimensional fixed effects are included, more than 95 percent of the variation in bilateral migration flows is absorbed. Geographic distance and the crossing of inter-municipal borders remain the dominant determinants, while most time-varying push and pull factors display limited explanatory power. Housing market variables show more consistent associations with flows than income or labor market indicators, but their effects remain modest. Environmental exposures, including flood and heatwave risks, are not associated with large-scale migration responses.

Overall, the gravity structure of mobility remains remarkably stable, including around the Covid-19 period. These findings suggest that rural development trajectories are shaped more by persistent territorial structures than by short-term shocks, with important implications for place-based regional policies.

Mr Simon Delage (p)

Territorial attractiveness has become a central concept in French spatial planning policy, yet existing measurement tools remain poorly adapted to local realities: most indices operate at the regional scale, focus on observed flows rather than intrinsic territorial assets, and systematically neglect the needs of existing residents in favour of capturing mobile populations. To address these gaps, this paper introduces the French Localities Attractiveness IndicatoR (FLAIR), a composite index covering all 34,842 municipalities of hexagonal France and Corsica. Conceptualizing attractiveness as the set of territorial assets that enable residents to emancipate and flourish, rather than competitive advantages designed to attract outsiders, the FLAIR aggregates 50 indicators from French open administrative data into six dimensions: Infrastructure and Services, Social Cohesion, Local Economy, Tourism, Housing, and Risk Exposure. The construction follows the OECD Handbook on Constructing Composite Indicators, with spatial smoothing using travel-time buffers calibrated to urban-rural typology, min-max normalization, and geometric mean aggregation to ensure partial non-compensability across dimensions. The findings reveal a continuous gradient of attractiveness rather than a binary territorial fracture, with intermediate urban municipalities outperforming both metropolitan centres and very low-density rural areas. The dimensions most differentiating high- and low-performing municipalities are social cohesion, residential environment, and service access, not economic performance, challenging dominant competitive narratives and pointing toward a reorientation of territorial policy around habitability.

Mr Pau Costa Capdevila (p)

In this study, we explore the determinants of internal mobility in Spain, focusing on ruralurban
disparities. Using microdata from the Estadística de Variaciones Residenciales (EVR)
from INE, we built a novel origin and destination matrix dataset on migration flows at the
municipal level for Spain. We exploit this database by elaborating an internal migration model
based on gravity analysis. We also use breakthrough data on income levels per capita at the
municipal level from the Atlas of Income Distribution elaborated by the Instituto Nacional
de Estadística (INE), which serves as a proxy for PIB per capita at the destination over the
period from to 2016-2021. We then evaluate the role of income, infrastructure, renewable
energy, and housing rent in shaping these flows using a Poisson Pseudo-Maximum Likelihood
(PPML) approach with extensive destination and origin-year fixed effects. Our findings show
that income per capita acts as a significant pull factor, whereas distance and housing rent act
as push components. We document substantial heterogeneity by urban-rural and immigrantnative
origins, which provides relevant insights for policies oriented toward reducing regional
disparities.