Vertical micro-segregation: is living in disadvantageous lower floors in Athens’ apartment blocks producing negative social effects?

Abstract

This article estimates the effect of vertical segregation – across floors of apartment buildings – on educational attainment in the metropolitan area of Athens. This context offers an opportunity to evaluate and discuss the complementarity between horizontal segregation (across neighbourhoods) and vertical segregation. Using census microdata, we observe that the share of individuals dropping out of school early is much higher for young residents of disadvantageous lower-floor apartments than for those living in upper floors. This gap remains significant after controlling for all relevant personal, household, and neighbourhood characteristics. This result suggests an effect of vertical segregation on educational outcomes in addition to neighbourhood effects and individuals’ socio-economic status. Moreover, the findings of this article corroborate the claim that the target for public policies to increase social mix is not enough to address inequalities at the local level.

Keywords:

• Vertical segregation
• spatial effects
• early school leavers
• gentrification
• apartment blocks
• Athens

Acknowledgements

This article was partly inspired by debates and research conducted in the ISTOPOL research project (2020–2022).

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This project has received funding from the Hellenic Foundation for Research & Innovation (HFRI) under grant agreement No 18162-30/10/2019.

Notes

1 The household reference person corresponds to the parent living with the child for sole parents. For children living with both parents, the education level of the father was kept.

2 The distinction between developed and developing countries is made according to the UN’s Human Development Index (UNDP, 2011, 127–130). We subsequently use the term ‘Greeks’ for Greeks and other nationals from developed countries. Greeks represent 97.3% of this category.

3 The coefficient associated with the unemployment rate was 0.048 when we only included the aforementioned variables and only this variable at the neighborhood level. A 1 pp increase in the unemployment rate of the neighborhood led to an expected increase in the log odds of being an early leaver of 0.048. We had exp(0.048*6) = 1.33.

Additional information

Notes on contributors

Thomas Maloutas

Thomas Maloutas is Professor at the Department of Geography at Harokopio University. He is the former Director of the National Centre for Social Research and was also Professor at the Department of Planning, University of Thessaly. Thomas Maloutas served as General Secretary for Research & Technology. His work is related to changing social structures in metropolitan areas with a focus on issues of segregation and gentrification related to housing and broader welfare regimes. His research and published work refer mainly to the South European urban context and especially to Athens.

Hugo Botton

Hugo Botton is a student at the ENS Paris-Saclay university, in the Department of Economics. He graduated from the master in Public Policy and Development (PPD) of the Paris School of Economics (PSE). He visited Harokopio University during a pre-doctoral year in 2020. His research mainly focuses on urban socio-spatial inequalities, their implications on people’s daily life and opportunities as well as the evaluation and the conception of public policies to limit these phenomena.